Speech generation system and method

Abstract

A sound generation system is disclosed. The sound generation system includes a housing with a first opening and a second opening. The housing defines an air passage between the first opening and the second opening. The sound generation system includes a movable member disposed within the housing. The movable member is configured to vibrate in response to air flowing through the air passage. The sound generation system includes a conversion module configured to convert vibrations of the movable member into an electrical signal. The sound generation system includes a speaker module configured to convert the electrical signal into sound and output the sound into the user's oral cavity. The sound generation system is configured such that, during use, air pressure at or near the second opening corresponds to air pressure within the user's oral cavity.

Description

Technical Field

[0001] The present invention generally relates to artificial voice generation. In a more specific example, the present invention relates to a method and system for generating voice.

Background Art

[0002] The pharynx, also known as the sound generator, is an organ used to generate the sounds that humans use in conversation. The pharynx houses the vocal cords, which are the source of human voice. When a healthy human speaks, the sound (or "voice") produced by the vocal cords in the pharynx enters the vocal tract, where the voice is filtered to produce speech (e.g., by controlling the movement of the tongue and lips). People whose larynx has been surgically removed (by laryngectomy) or bypassed (by tracheotomy) may still be able to control the vocal tract, but lack the vocal cords needed to produce voice. Therefore, such people are unable to produce voice or their ability to do so is suppressed without artificial assistance.

[0003] An example of artificial assistance is the electric larynx, which is a portable device that a laryngectomy patient presses against the skin of the neck or face to speak. This device functions as an artificial sound source by inducing vibrations in the vocal tract, so that the person can vocalize the vibrations by controlling the movement of the tongue and lips. However, the voice produced by the electric larynx tends to sound robotic, and generally, it is inconvenient that the device has to be manually operated while the person is speaking.

[0004] A tracheoesophageal voice prosthesis (TEP) is another type of voice prosthesis (artificial aid) that has traditionally been used by patients who have undergone laryngectomy. During laryngectomy, a permanent opening known as a stoma is created in the patient's neck for breathing. As a result, the patient's trachea is no longer in communication with the vocal tract, and air from the lungs can no longer enter the vocal tract by passing through the stoma. A TEP is a plastic valve that is surgically inserted into the throat between the trachea and esophagus. The TEP allows air from the lungs to re-enter the esophagus, from there through the throat, vibrating the tissue inside the throat and consequently producing sound (similar to how belching produces sound). Although the resulting speech is clear, TEP is an early solution and has several significant drawbacks. For example, TEP is highly invasive, carries a biohazard of infection and swallowing, and the voice produced is limited to a hoarse whisper.

[0005] Novel or improved systems and / or methods are needed for generating sound or generating airflow through the human vocal tract.

[0006] Any reference in this Specified Publication to any prior publication (or information derived from prior publication) or any publicly known matter shall not be construed as an agreement, acknowledgment, or any form of suggestion that the prior publication (or information derived from prior publication) or publicly known matter constitutes part of the common technical knowledge in the relevant field of this Specified Publication. [Overview of the project]

[0007] This “Summary of the Invention” is provided in a concise form to introduce the selection of concepts that will be further described later in the “Modes for Carrying Out the Invention.” This “Summary of the Invention” is not intended to identify the main or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

[0008] In one embodiment, a sound generation system is provided, comprising: a housing having a first opening and a second opening, the housing defining an air passage between the first and second openings; a movable member disposed within the housing and configured to vibrate in response to air flowing through the air passage; a conversion module configured to convert the vibration of the movable member into an electrical signal; and a speaker module configured to convert the electrical signal into sound and output sound into the user's oral cavity, wherein during use, the air pressure at or near the second opening corresponds to the air pressure inside the user's oral cavity.

[0009] In certain embodiments, the flow of air within the air passage is due to the difference in air pressure between the first opening and the second opening.

[0010] In certain embodiments, the air pressure at or near the first opening corresponds to the air pressure at the user's neck.

[0011] In certain embodiments, the air pressure at or near the first opening corresponds to the air pressure in the user's neck stoma.

[0012] In a particular embodiment, the first opening is configured to communicate with the user's neck stoma such that the air pressure at the first opening corresponds to the air pressure at the user's neck stoma.

[0013] In certain embodiments, the second opening is configured to communicate with the user's oral cavity such that the air pressure in the second opening corresponds to the air pressure in the user's oral cavity.

[0014] In certain embodiments, the sound generation system further comprises a tube having a first end connected to a housing around a second opening, the second end of which is configured to be inserted into the user's oral cavity.

[0015] In certain embodiments, the housing and the pipe together define an air passage between the first opening of the housing and the second end of the pipe.

[0016] In certain embodiments, the speaker module is located within a housing and is configured to output sound through a second opening.

[0017] In certain embodiments, the housing is configured to connect to an airflow source, which is configured to generate an airflow and output the airflow into the user's oral cavity.

[0018] In certain embodiments, the airflow source is the user's neck stoma, and the airflow is the airflow of breathing expelled from the user through the neck stoma.

[0019] In certain embodiments, the air source is an air pump.

[0020] In a particular embodiment, the voice generation system further includes a pressure sensing module that is attached to the user's neck and configured to detect air pressure in the user's neck stoma; an air pump located within the housing and configured to generate an airflow that moves along an air passage from a first opening to a second opening; and a controller configured to control the air pump based on the detected air pressure in the neck stoma.

[0021] In certain embodiments, the flow of air within the air passage is due to the difference between the air pressure within the air passage and the air pressure at the second opening.

[0022] In certain embodiments, the second opening is configured to communicate with the user's oral cavity.

[0023] In certain embodiments, the sound generation system further comprises a tube having a first end connected to a housing around a second opening, the second end of which is configured to be inserted into the user's oral cavity.

[0024] In certain embodiments, both the housing and the tube define an air passage between a first opening of the housing and a second end of the tube.

[0025] In certain embodiments, the speaker module is disposed within the housing, and the speaker module is configured to output sound through the second opening.

[0026] In certain embodiments, the air pump is configured to generate an air pressure within the air passage corresponding to a detected air pressure at the neck stoma.

[0027] In certain embodiments, the housing is configured to be fixed to the user's auricle.

[0028] In certain embodiments, the sound generation system is configured to be used hands-free.

[0029] According to another exemplary aspect, a sound generation system is provided, the system comprising: a housing having a first opening and a second opening, the housing defining an air passage between the first opening and the second opening, the first opening being configured to communicate with a neck stoma of a user; a movable member disposed within the housing and configured to vibrate in response to air flowing through the air passage; an adjuster disposed within the housing and configured to control the air flow within the air passage; a conversion module configured to convert the vibration of the movable member into an electrical signal; a speaker module configured to convert the electrical signal generated by the conversion module into sound and output the sound into the user's oral cavity; an air pump configured to generate an air flow into the user's oral cavity; a pressure detection module configured to detect an air pressure within the oral cavity; and a controller configured to control the adjuster based on the detected air pressure within the oral cavity.

[0030] In certain embodiments, the housing is configured to attach to the user's neck such that the first opening connects to the neck stoma.

[0031] In certain embodiments, the housing is attached to a neck harness configured to fit around the back of the user's neck.

[0032] In certain embodiments, the sound generation system further includes a flow detection module disposed within the housing and configured to detect an air flow within the air passage,

[0033] and a second controller configured to control the air pump based on the detected air flow within the air passage, wherein the air pump is configured to generate an air flow to the oral cavity corresponding to the detected air flow within the air passage.

[0034] In certain embodiments, the air pump and the speaker module are disposed within a second housing configured to be fixed to the user's auricle, the second housing includes an opening configured to communicate with the user's oral cavity, and the speaker module is configured to output sound through the opening of the second housing.

[0035] In certain embodiments, the sound generation system further includes a tube having a first end connected to the opening of the second housing, the second end of the tube is configured to be inserted into the user's oral cavity, and the pressure detection module is attached to the second end of the tube.

[0036] In certain embodiments, the air pump, the speaker module, and the pressure detection module are attached to a denture unit configured to be fixed to the user's oral cavity.

[0037] In certain embodiments, the regulator is a further air pump configured to generate an air pressure within the air passage corresponding to the detected air pressure within the oral cavity.

[0038] In certain embodiments, the regulator is an air valve configured to control the flow of air through a second opening.

[0039] In certain embodiments, the pressure sensing module communicates wirelessly with the controller, and the conversion module communicates wirelessly with the speaker module.

[0040] In certain embodiments, the voice generation system further comprises a processing system.

[0041] In certain embodiments, the processing system is configured to reduce sound interference received by the conversion module.

[0042] In a particular embodiment, the processing system is • A conversion module configured to convert vibrations of a movable member, and / or • At least one interference converter module for receiving sound interference, It is configured to reduce sound interference based on at least one electrical signal received from it.

[0043] In certain embodiments, sound interference in the conversion module includes sound output by the speaker module and / or phonations generated when the sound output by the speaker module is adjusted by the user's mouth movements.

[0044] In another embodiment, a sound generation system is provided, comprising: a housing having a first opening and a second opening, the housing defining an air passage between the first opening and the second opening; a movable member disposed within the housing and configured to vibrate in response to air flowing through the air passage; a conversion module configured to convert the vibration of the movable member into an electrical signal; a speaker module configured to convert the electrical signal into sound and output the sound into the user's oral cavity; and a processing system configured to reduce sound interference received by the conversion module.

[0045] In a particular embodiment, the processing system is configured to reduce sound interference based on at least one electrical signal received from: a conversion module configured to convert vibrations of a movable member, and / or at least one interference conversion module for receiving sound interference.

[0046] In certain embodiments, the processing system is configured to reduce sound interference by performing feedforward suppression.

[0047] In certain embodiments, feedforward suppression includes one or more of the following: gain reduction, frequency-selective noise reduction, notch filtering, phase modulation, and frequency shifting.

[0048] In certain embodiments, the processing system is configured to reduce sonic interference by performing adaptive feedback cancellation or residual feedback suppression.

[0049] In certain embodiments, adaptive feedback cancellation includes adaptive filtering.

[0050] In certain embodiments, adaptive filtering is performed using an FIR filter.

[0051] In certain embodiments, adaptive filtering is performed using a (real-time) normalized least squares mean (NLMS) algorithm.

[0052] In certain embodiments, the voice generation system further comprises at least one interference conversion module for receiving sound interference.

[0053] In certain embodiments, sound interference in the conversion module includes sound output by the speaker module and / or phonations generated when the sound output by the speaker module is adjusted by the user's mouth movements.

[0054] In certain embodiments, the sound generation system further includes a first interference converter module configured to convert the sound output by the speaker module into an electrical signal for input to a processing system.

[0055] In certain embodiments, the speech generation system further includes a second interference converter module configured to convert the utterances produced when the sound output by the speaker module is adjusted by the user's mouth movements into electrical signals for input to a processing system.

[0056] In a particular embodiment, the processing system outputs at least one electrical signal to reduce sound interference.

[0057] In a particular embodiment, the interference conversion module is • Microphone • Piezoelectric converter • Magnetic pickup converter ·Accelerometer • Voice sensor • Vibration sensor It includes one of the following.

[0058] In certain embodiments, the processing system is configured to process electrical signals generated by a conversion module.

[0059] In certain embodiments, the processing system comprises hardware and software configured to improve the volume or quality of audio encoded by electrical signals.

[0060] In certain embodiments, the processing system includes AI speech conversion software for improving the quality of speech encoded by electrical signals.

[0061] In certain embodiments, the conversion module includes a microphone.

[0062] In certain embodiments, the conversion module includes a piezoelectric converter.

[0063] In certain embodiments, the conversion module includes a magnetic pickup converter.

[0064] In certain embodiments, the speaker module comprises a loudspeaker array.

[0065] In a particular embodiment, the conversion module is • Pressure sensor • Sound com • Vibration detection sensor ·Accelerometer It includes one of the following.

[0066] In certain embodiments, the movable member includes a membrane.

[0067] In certain embodiments, the movable member comprises a plurality of membranes having different degrees of complexity.

[0068] In another embodiment, a sound generation system is provided, comprising: a housing having a first opening and a second opening, the housing defining an air passage between the first opening and the second opening; a movable member disposed within the housing and configured to vibrate in response to air flowing through the air passage; a conversion module configured to convert the vibration of the movable member into an electrical signal; and a speaker module configured to convert the electrical signal into sound and output sound into the user's oral cavity, wherein the output sound is based on the air pressure in the user's oral cavity.

[0069] In certain embodiments, the voice generation system is configured such that, during use, the air pressure at or near the second opening corresponds to the air pressure inside the user's mouth.

[0070] In another embodiment, a method for generating sound is provided, which comprises the steps of: providing a movable member that communicates with (interacts with) an airflow corresponding to the airflow of the user's breathing; detecting vibrations of the movable member in accordance with the airflow; generating sound originating from the detected vibrations using one or more electric speakers; and supplying airflow and sound to the user's oral cavity.

[0071] In a particular embodiment, the movable member is provided within an air passage, which is subjected to a first air pressure at a first end of the air passage and a second air pressure at a second end of the air passage.

[0072] In certain embodiments, the second air pressure corresponds to the air pressure inside the user's mouth.

[0073] In certain embodiments, the first air pressure corresponds to the air pressure of the user's neck stoma.

[0074] In certain embodiments, the method further, A process for detecting air pressure and / or airflow in the user's neck stoma, A process of generating the first air pressure and airflow of the neck stoma using an air pump, It is equipped with.

[0075] In certain embodiments, the method further,

[0076] A process to detect the air pressure inside the user's mouth,

[0077] The process includes a step of generating a second air pressure using an airflow control element.

[0078] In certain embodiments, the airflow control element is an air pump.

[0079] In certain embodiments, the airflow control element is an air valve.

[0080] In certain embodiments, the airflow control element is ·Suction device Actuator It is one of the following.

[0081] In certain embodiments, the method further comprises the step of guiding the airflow of the user's breath within an air passage, wherein the airflow is the guided airflow of the breath.

[0082] In a particular embodiment, the method further comprises the steps of: using a converter to convert the vibration of a movable member into an electrical signal; processing the electrical signal; and supplying the processed electrical signal to one or more electrical speakers in order to generate sound.

[0083] In certain embodiments, the process of processing an electrical signal includes the process of improving the volume or quality of the sound encoded by the electrical signal.

[0084] In certain embodiments, the method further comprises the step of reducing sound interference generated by one or more electric speakers.

[0085] In one embodiment, a sound generation system is provided, comprising: a housing having a first opening and a second opening, the housing defining an air passage between the first opening and the second opening; a movable member disposed within the housing and configured to vibrate in response to air flowing through the air passage; a conversion module configured to convert the vibration of the movable member (or sound produced by the vibration) into an electrical signal; and a speaker module configured to convert the electrical signal into sound and output the sound into the user's oral cavity.

[0086] In certain embodiments, the flow of air within the air passage is due to the difference in air pressure between the first opening and the second opening.

[0087] In certain embodiments, the first opening is configured to communicate with the user's neck stoma.

[0088] In certain embodiments, the housing is configured to adhere to the user's neck or chest such that the first opening connects to a neck stoma.

[0089] In certain embodiments, the second opening is configured to communicate with the user's oral cavity.

[0090] In certain embodiments, the sound generation system further comprises a tube having a first end connected to a housing around a second opening, the second end of which is configured to be inserted into the user's oral cavity.

[0091] In certain embodiments, the housing and the pipe together define an air passage between the first opening of the housing and the second end of the pipe.

[0092] In certain embodiments, the speaker module is located within a housing and is configured to output sound (i.e., speech) through a second opening.

[0093] In certain embodiments, the housing is configured to connect to an airflow source, which is configured to generate an airflow and output the airflow into the user's oral cavity.

[0094] In certain embodiments, the airflow source is the user's neck stoma, and the airflow is the airflow of breathing expelled from the user through the neck stoma.

[0095] In certain embodiments, the air source is an air pump.

[0096] In a particular embodiment, the voice generation system further includes a pressure sensing module that is attached to the user's neck and configured to detect air pressure in the user's neck stoma; an air pump located within the housing and configured to generate an airflow that moves along an air passage from a first opening to a second opening; and a controller configured to control the air pump based on the detected air pressure in the neck stoma (monitored by the pressure sensing module).

[0097] In certain embodiments, the flow of air within the air passage is due to the difference between the air pressure within the air passage and the air pressure at the second opening.

[0098] In certain embodiments, the second opening is configured to communicate with the user's oral cavity.

[0099] In certain embodiments, the sound generation system further comprises a tube having a first end connected to a housing around a second opening, the second end of which is configured to be inserted into the user's oral cavity.

[0100] In certain embodiments, the housing and the pipe together define an air passage between the first opening of the housing and the second end of the pipe.

[0101] In certain embodiments, the speaker module is located within a housing and is configured to output sound through a second opening.

[0102] In certain embodiments, the air pump is configured to generate air pressure in the air passage that corresponds to the detected air pressure in the neck stoma.

[0103] In certain embodiments, the housing is configured to be fixed to the user's auricle.

[0104] In another embodiment, a sound generation system is provided, comprising: a housing having a first opening and a second opening, the housing defining an air passage between the first opening and the second opening, the first opening being configured to communicate with the user's neck stoma; a movable member disposed within the housing and configured to vibrate in response to air flowing through the air passage; a conversion module configured to convert the vibration of the movable member into an electrical signal; a speaker module configured to convert the electrical signal into sound and output the sound into the user's oral cavity; a first air pump configured to generate air pressure at the second opening in the air passage; a pressure detection module configured to detect the air pressure in the user's oral cavity; a controller configured to control the first air pump based on the detected air pressure in the oral cavity; an airflow detection module configured to detect the airflow of the user's stoma; and a second air pump configured to generate airflow into the user's oral cavity based on the detected airflow of the neck stoma.

[0105] In another embodiment, a sound generation system is provided, comprising: a housing having a first opening and a second opening, the housing defining an air passage between the first opening and the second opening, the first opening being configured to communicate with the user's neck stoma; a movable member disposed within the housing and configured to vibrate in response to air flowing through the air passage; a regulator disposed within the housing and configured to control the airflow in the air passage; a conversion module configured to convert the vibration of the movable member into an electrical signal; a speaker module configured to convert the electrical signal generated by the conversion module into sound and output the sound into the user's oral cavity; an air pump configured to generate an airflow into the user's oral cavity; a pressure sensing module configured to detect the air pressure in the user's oral cavity; and a controller configured to control the regulator based on the detected air pressure in the oral cavity.

[0106] In certain embodiments, the housing is configured to adhere to the user's neck such that the first opening connects to a neck stoma.

[0107] In certain embodiments, the housing is attached to a neck harness configured to fit around the back of the user's neck.

[0108] In certain embodiments, the voice generation system further comprises a flow detection module located within the housing and configured to detect airflow in an air passage (e.g., stoma airflow), and a second controller configured to control an air pump based on the detected airflow in the air passage, the air pump being configured to generate an airflow to the oral cavity corresponding to the detected airflow in the air passage.

[0109] In certain embodiments, the air pump and speaker module are housed within a second housing configured to be fixed to the user's ear, the second housing having an opening configured to communicate with the user's oral cavity, and the speaker module is configured to output sound through the opening in the second housing.

[0110] In certain embodiments, the voice generation system further comprises a tube having a first end connected to an opening in a second housing, the second end of which is configured to be inserted into the user's oral cavity, and a pressure sensing module attached to the second end of the tube.

[0111] In certain embodiments, the air pump, speaker module, and pressure sensing module are attached to a denture unit configured to be fixed in the user's mouth.

[0112] In certain embodiments, the regulator is a further air pump configured to generate an air pressure in the air passage corresponding to the air pressure detected in the oral cavity. In certain embodiments, the regulator is an air valve configured to control the airflow through the second opening.

[0113] In certain embodiments, the pressure sensing module communicates wirelessly with the controller, and the conversion module communicates wirelessly with the speaker module.

[0114] In certain embodiments, the voice generation system further comprises a processing system configured to process the electrical signals generated by the conversion module.

[0115] In certain embodiments, the processing system comprises hardware and software configured to improve the volume or quality of audio encoded by electrical signals.

[0116] In certain embodiments, the processing system includes artificial intelligence (AI) speech conversion software for improving the quality of speech encoded by electrical signals.

[0117] In certain embodiments, the software is implemented as a trainable artificial intelligence (AI) algorithm.

[0118] In certain embodiments, the conversion module includes a microphone.

[0119] In certain embodiments, the conversion module includes a piezoelectric converter.

[0120] In certain embodiments, the conversion module includes a magnetic pickup converter.

[0121] In certain embodiments, the speaker module comprises a microspeaker or a loudspeaker array.

[0122] In certain embodiments, the movable member includes a membrane.

[0123] In certain embodiments, the movable member comprises a plurality of membranes having different degrees of complexity.

[0124] In certain embodiments, the movable member is a silicone model of the vocal cords.

[0125] In another embodiment, a method for generating sound is provided, which comprises the steps of: providing a movable member that communicates with (interacts with) an airflow corresponding to the airflow of the user's breathing; detecting vibrations of the movable member in accordance with the airflow; generating sound originating from the detected vibrations using one or more electric speakers; and supplying airflow and sound to the user's oral cavity.

[0126] In a particular embodiment, the movable member is provided within an air passage, which is subjected to a first air pressure at a first end of the air passage and a second air pressure at a second end of the air passage.

[0127] In a particular embodiment, the first air pressure corresponds to the air pressure in the user's neck stoma, and the second air pressure corresponds to the air pressure in the user's oral cavity.

[0128] In a particular embodiment, the method further comprises the steps of detecting air pressure and / or airflow in the user's neck stoma and generating a first air pressure and airflow in the neck stoma using an air pump.

[0129] In certain embodiments, the method further comprises the steps of detecting the air pressure in the user's oral cavity and generating a second air pressure using a regulator or airflow control element. In certain embodiments, the airflow control element is an air pump. In certain embodiments, the airflow control element is an air valve.

[0130] In certain embodiments, the method further comprises the step of guiding the airflow of the user's breath within an air passage, wherein the airflow is the guided airflow of the breath.

[0131] In a particular embodiment, the method further comprises the steps of: using a converter to convert the vibration of a movable member into an electrical signal; processing the electrical signal; and supplying the processed electrical signal to one or more electrical speakers in order to generate sound.

[0132] In certain embodiments, the process of processing an electrical signal includes the process of improving the volume or quality of the sound encoded by the electrical signal.

[0133] Other aspects, features, and advantages will become apparent from the following “Modes for Carrying Out the Invention,” which are part of this disclosure and are described in conjunction with the accompanying drawings illustrating the principles of various embodiments. [Brief explanation of the drawing]

[0134] Examples of embodiments will become apparent from the following description (given merely as examples) of at least one non-limiting embodiment described in conjunction with the attached drawings.

[0135] [Figure 1A] A block diagram showing an example of a voice generation system. [Figure 1B] A block diagram showing an example of a voice generation system.

[0136] [Figure 2] A block diagram showing another example of a voice generation system.

[0137] [Figure 3] A block diagram showing another example of a voice generation system.

[0138] [Figure 4] A front perspective view showing the voice generation system of Figure 3 as worn by the user.

[0139] [Figure 5] A side perspective view showing the voice generation system shown in Figure 3, worn by the user.

[0140] [Figure 6] Figure 3 is a perspective view showing the housing and tubing of the sound generation system.

[0141] [Figure 7] Figure 3 is a perspective view showing the housing of the sound generation system.

[0142] [Figure 8] Figure 3 is an exploded view showing the sound generation system.

[0143] [Figure 9] Figure 3 shows an example of a movable member of the sound generation system.

[0144] [Figure 10] Figure 3 shows an example of adhesive used to attach the pressure sensing module of the voice generation system to the user's neck.

[0145] [Figure 11] Figure 3 shows an example of a pressure sensing module in the voice generation system when it is attached to the user's neck.

[0146] [Figure 12] A block diagram showing another example of a voice generation system.

[0147] [Figure 13] Figure 12 shows a voice generation system in which the housing is supported by a neck harness.

[0148] [Figure 14] A block diagram showing another example of a voice generation system.

[0149] [Figure 15] A block diagram showing the mouse plate and pressure / airflow sensing module of the wireless voice generation system in Figure 14.

[0150] [Figure 16] A diagram illustrating an example of a method for generating sound.

[0151] [Figure 17] A diagram illustrating an example of sound interference in a sound generation system.

[0152] [Figure 18] A block diagram illustrating a method for eliminating sound interference in a sound generation system. [Modes for carrying out the invention]

[0153] The following embodiments are given only as examples and are described to provide a more accurate understanding of the subject matter of the embodiments. In the drawings incorporated to illustrate the features of the embodiments, similar reference numerals are used to identify similar parts throughout the drawings.

[0154] This invention describes an electronically respiration-driven voice generation system for generating voice and / or airflow components for human use. The system functions as an artificial voice (and / or airflow) source for a person to replace or enhance the voice-generating function of the vocal cords. Thus, this system may be called an “air-based bionic voice” source.

[0155] A "breath-driven" voice generation system generates sound in response to fluctuations in respiratory pressure or airflow. The system may provide access to these fluctuations by monitoring air pressure or airflow in front of the stoma and / or in the mouth of laryngectomized patients.

[0156] Here, a general embodiment of the voice generation system will be described with reference to Figures 1A and 1B.

[0157] Figure 1A shows the voice generation system 100. The voice generation system 100 comprises a housing 110 having a first opening 112 and a second opening 114. The housing 110 defines an air passage, air flow path, or air duct 116 between the first opening 112 and the second opening 114.

[0158] The sound generation system 100 further includes a movable member 120 located within the housing 110 and configured to vibrate in response to airflow in the air passage 116 and / or in response to fluctuations in air pressure at the first opening 112 and the second opening 114 or within the air passage 116. In some examples, the movable member 120 is located within the air passage 116. In some examples, the movable member 120 extends across the air passage 116 in a direction perpendicular to the airflow within the air passage 116.

[0159] In some examples, the movable member 120 comprises a membrane. In some examples, the movable member 120 comprises a physical structure configured to vibrate in response to air flowing through the air passage 116. The physical structure of the movable member 120 may represent or correspond to a mechanical model of the human vocal cords with varying degrees of complexity.

[0160] The voice generation system 100 further comprises a conversion module 130. The conversion module may be located within the housing 110. The conversion module 130 is configured to convert vibrations of the movable member 120 into electrical signals. In some examples, the conversion module 130 is physically coupled to the movable member 120. In other examples, the conversion module 130 is operably coupled to the movable member 120. In some examples, the conversion module 130 comprises one or more microphones or microphone arrays. In some examples, the conversion module 130 comprises one or more piezoelectric transducers or magnetic pickup transducers, which have the advantage of reducing or avoiding voice interference from external sound sources. In some embodiments, the conversion module 130 may include sensors. These may be pressure, sound, or vibration detection sensors. In some embodiments, the conversion module 130 may include accelerometers. In this specification, the term “conversion module” (including interference conversion module) may include any example mentioned. It may also include voice detection sensors, sound sensors, vibration sensors, etc.

[0161] The voice generation system 100 further comprises a speaker module 140. In some examples, the speaker module is located within the housing 110. In other examples, the speaker module is not located within the housing 110. The speaker module 140 is configured to convert the electrical signal generated by the conversion module 130 into sound and output the sound to the user's mouth. In some examples, the speaker module 140 comprises one or more loudspeakers or loudspeaker arrays. In some examples, the frequency response of the loudspeakers is flat (e.g., with fluctuations of less than 3 dB) within the frequency range of a human voice source (e.g., between approximately 50 Hz and approximately 1000 Hz).

[0162] In some examples, the conversion module 130 and the speaker module 140 communicate (such as via wired or wireless communication) to enable the speaker module 140 to access the electrical signals generated by the conversion module 130. In other examples, each of the conversion module 130 and the speaker module 140 communicates with one or more processing systems (such as the audio processing system 150 described later) that receive the electrical signals from the conversion module 130, optionally process or enhance the electrical signals (i.e., audio signals), and then supplies the audio as electrical signals to the speaker module 140 for conversion into sound.

[0163] In some embodiments, the voice generation system 100 may be configured to, during use, correlate the air pressure at or near the first opening 112 with the air pressure in the user's neck (e.g., the user's neck stoma) (or the air pressure in a respiratory airway such as the trachea). The air pressure at or near the first opening 112 is considered to correspond with the air pressure in the user's neck (stoma) if the first opening is in communication with the neck (stoma), or if the air pressure at or near the first opening is artificially generated as an air pressure that corresponds to (e.g., substantially the same as or substantially proportional to) the air pressure in the user's neck (stoma).

[0164] In some embodiments, it is desirable that the voice generation system 100 generates output sound (e.g., from a speaker module) based on the air pressure in the user's oral cavity or oral canal. This can be done as follows: • Second opening 114 or the air pressure in the vicinity thereof is made to correspond to the air pressure in the user's mouth or oral canal, and / or, For example, sound is generated based on the air pressure being monitored in the oral cavity or oral canal, such as by monitoring fluctuations in air pressure within the oral cavity.

[0165] In some embodiments, the air pressure at or near the second opening 114 may be made to correspond to the air pressure in the user's oral cavity or oral canal. The air pressure at or near the second opening 114 is considered to correspond to the air pressure in the user's oral cavity or oral canal if the second opening is in communication with the oral cavity or oral canal, or if the air pressure at or near the second opening is artificially generated to correspond to the air pressure in the user's oral cavity or oral canal (e.g., substantially the same or substantially proportional).

[0166] The advantage of correlating the air pressure at or near the second opening with the oral cavity or oral canal is to improve the quality of the sound output from the speaker module 140, and consequently, the quality of the voice output from the user. From experiments, the inventors found that when the air pressure at or near the second opening 114 is not correlated with the air pressure in the user's oral cavity or oral canal, the output phonation becomes monotonous. The inventors also found that it is desirable to be able to control the transition of the voice generation system between voice mode and silent mode in real time. From experiments, the inventors found that if the air pressure at or near the second opening is not correlated with the oral cavity or oral canal, the movable member will constantly vibrate if the person continues to exhale while phoning, and unless it is correlated, such control will be impaired. More generally, similar advantages apply when the output sound generated (e.g., from a speaker module) is based on the air pressure in the user's oral cavity or oral canal.

[0167] Furthermore, the advantages discussed regarding vocal quality also apply when the voice generation system produces output sound based on the air pressure inside the user's mouth.

[0168] Furthermore, in addition to generating output sound (e.g., from a speaker module) based on the air pressure in the user's oral cavity or oral canal (such as matching the air pressure at or near the second opening to the oral cavity or oral canal), the advantages discussed regarding speech quality become more apparent when the air pressure at or near the first opening is matched to the air pressure in the user's neck (e.g., the user's neck stoma). Therefore, it is preferable that the speech generation system 100 be configured to, when in use, match the air pressure at or near the second opening 114 to the air pressure in the user's oral cavity or oral canal, while simultaneously matching the air pressure at or near the first opening 112 to the air pressure in the user's neck (e.g., a neck stoma) (or the air pressure in a respiratory airway such as the trachea). The same advantages apply if the voice generation system 100 is configured such that, when in use, the output sound generated (e.g., from a speaker module) is based on the air pressure in the user's oral cavity or oral canal, and at the same time, the air pressure at or near the first opening 112 can be made to correspond to the air pressure in the user's neck (e.g., a neck stoma) (or the air pressure in a respiratory airway such as the trachea).

[0169] Since airflow (which may be referred to as “airflow” throughout this specification) is caused by differences in air pressure, a situation in which the airflow at or near the first end of the housing corresponds to the airflow in the neck (stoma) can be considered as a situation in which the air pressure at or near the first end of the housing corresponds to the air pressure in the neck (stoma). Similarly, a situation in which the airflow at or near the second end of the housing corresponds to the airflow in the oral cavity or oral canal can be considered as a situation in which the air pressure at or near the second end of the housing corresponds to the air pressure in the oral cavity or oral canal canal.

[0170] Therefore, the difference in air pressure between the trachea and the oral cavity (in the stoma) can facilitate airflow in the passage 116, causing vibration of the movable member 120. The vibration of the movable member 120 is converted into one or more electrical signals, which are then sent to a speaker module to synthesize speech in the user's oral cavity. By shaping, filtering, or modifying the speech (or sound) generated by the speaker module using the movements of the lips, tongue, and vocal tract, the user can produce speech. Thus, the speech generation system 100 may be called an “air-based bionic speech” source.

[0171] The speech generation systems and methods described herein can generate very high-quality speech for persons who have lost or damaged their larynx (vocal cords). The person uses the speech generation system as an artificial or augmented speech source to generate speech in the oral cavity and modulates this speech by moving the muscles of the mouth and lips in a natural speaking manner. The speech generation system and methods may provide automatic, real-time control of the beginning and end of speech when the person speaks using the system. Without this beginning / end control, voiceless phonemes would be produced, and vice versa, negatively impacting the clarity of the resulting speech.

[0172] In some cases, the beginning or end of the voice emitted by the speaker module 140 is controlled using variations in oral and stoma air pressure (e.g., as described in "A pneumatic Bionic Voice prosthesis-Pre-clinical trials of controlling the voice onset and offset" by Ahmadi Farzaneh Ahmadi et al., PloS one 13.2(2018):e0192257). Controlling the beginning and end of the voice using changes in breathing can help avoid undesirable voice feedback between the speaker module 140 and the conversion module 130 and reduce or avoid voice interference from external sound sources. This will be discussed in more detail later.

[0173] The voice generation system 100 does not require surgical placement and is non-invasive. Furthermore, since the voice generation system 100 can operate using the difference in air pressure between the stoma and the oral cavity, the user can speak without requiring manual operation.

[0174] In some examples, the air pressure / airflow at or near the first opening 112 is set or generated naturally, for example, by positioning the housing 110 so that the first opening 112 is connected to or communicates with the user's neck stoma (an opening in the front of the user's neck that communicates with the trachea so that the user inhales and exhales air through the stoma). In other examples, the air pressure at or near the first opening 112 is set or generated artificially, for example, by providing the housing 110 with an air pump that generates a pressure corresponding to the air pressure at the neck stoma. In some examples, a pressure sensor detects the air pressure or airflow at the neck stoma, and the air pump is operated based on the detected air pressure / airflow.

[0175] In some examples, the air pressure in or near the second opening 114 is set or generated naturally, for example, by positioning the housing 110 so that the second opening 114 connects to or communicates with the user's oral cavity. In other examples, the air pressure in or near the second opening 114 is set or generated artificially, for example, by providing the housing 110 with an air pump that generates pressure corresponding to the air pressure in the oral cavity or vocal tract. In some examples, a pressure sensor detects the air pressure in the oral cavity, and the air pump is operated based on the detected air pressure.

[0176] An air micropump or any other airflow source may be used to generate a given air pressure and / or airflow. In some examples, the airflow source comprises a microblower or an array of air nozzles. In some examples, the air pump or airflow source can produce an airflow rate that drives human voice breathing, the flow rate being between approximately 5 liters / min and approximately 10 liters / min or any other volumetric flow rate.

[0177] In addition to providing sound (or speech), the speech generation system 100 may further provide airflow into the user's oral cavity to facilitate the generation of voiceless phonemes (fricatives (e.g., / s / or / f)) in speech. In some examples, the airflow provided into the oral cavity is similar to, copies, mimics, or reflects the airflow passing through the air passage 116 due to the pressure difference between the first opening 112 and the second opening 114. In some examples, the airflow provided into the oral cavity is the same airflow passing through the air passage 116 due to the pressure difference between the first opening 112 and the second opening 114. In some examples, the housing 110 is configured such that the air passage 116 communicates with the user's oral cavity in order to supply airflow into the oral cavity.

[0178] In other examples, the airflow supplied to the user's oral cavity is different from and separate from the airflow passing through the air passage 116. In some examples, the voice generation system 100 further includes an air supply unit (such as an air pump) that communicates with the user's oral cavity and is configured to generate a second airflow to the oral cavity (other than the airflow passing through the air passage 116).

[0179] The voice generation system 100 may further include a processing system 150. The processing system 150 may be configured to process the electrical signals generated by the conversion module. The processing system 150 may be configured to process the electrical signals generated by the conversion module 130 in order to synthesize higher quality or higher amplitude voice. The processing system 150 may be provided as part of the conversion module 130 and the speaker module 140, or it may be a separate module distinct from these two modules. In this specification, the processing system 150, and the more general term “processing system,” may refer to one or more processors that can operate in conjunction or independently. That is, various functions or parts thereof performed by the processing system may be performed on a common processor or on multiple processors.

[0180] In some examples, the processing system 150 may run or execute voice enhancement software to improve the quality of the voice produced by the voice generation system 100. The processing system 150 may be configured to enhance the voice by sending it to a more advanced statistical voice conversion module based on artificial intelligence (AI) through amplification, spectral enhancement, and / or pitch shifting. The AI ​​algorithm may be trained to improve the quality of the voice or even to imitate the user's natural voice. In some examples, the AI ​​module is a statistical voice conversion module (e.g., "Voice conversion based on maximum-likelihood estimation of spectral parameter trajectory" by Toda, Tomoki, Alan W. Black, and Keiichi Tokuda, IEEE Transactions on Audio, Speech, and Language Processing 15.8(2007)2222-2235) configured to convert the voice generated by the movable member 120 and picked up by the conversion module 130 in the voice generation system 100 to sound more natural. The AI ​​algorithm may divide the speech generated by the conversion module 130 into three sets of parameters (pitch (f0), spectral information, and aperiodicity). The algorithm may then use a trained AI statistical engine to convert these parameters into a more natural-sounding speech. Next, the algorithm may combine the estimated parameters (pitch (f0), spectral information, and aperiodicity) in a vocoder to synthesize an enhanced, more natural-sounding speech. The algorithm may be trained with the user's natural speech before the laryngectomy so that the algorithm can learn to convert the speech generated by the movable member 120 into the user's natural speech.

[0181] The AI ​​speech conversion module may be trained with a mechanical or electromechanical pneumatic speech source equipped with a more complexly shaped movable member 120 used by laryngectomy patients. By using a more complex movable member, such as a mechanical model of the vocal cords containing a silicone membrane similar to the physical attributes of the vocal cords, the speech produced by the speech generation system 100 may sound closer to natural speech. However, driving a more complex pneumatic mechanical model requires more breathing effort from the patient, making it difficult for certain patients to drive the source. Therefore, a sample of a wearable, pneumatic complex mechanical vocal cord model may be constructed, for example, with a silicone model of the vocal cords, and may be used by certain able patients to produce speech during speech. Statistical speech conversion AI software may be trained to convert speech (or underlying breath) produced by the speech generation system 100, which has a simple membrane as the movable member 120, into speech that can be produced using these more complex and more natural-sounding vocal cord models (membranes). A trained AI module can convert simple (easy to drive) membrane sounds (or the underlying breaths) into more natural-sounding sounds from selected, more complex membranes.

[0182] In several examples, a breath-driven electromechanical voice generation system is provided. The voice generation system comprises a housing having a first opening and a second opening. The housing defines an air passage between the first and second openings. The voice generation system further comprises a movable member located within the housing and configured to vibrate in response to air flowing through the air passage. The voice generation system further comprises a conversion module configured to convert the vibration of the movable member into an electrical signal. The voice generation system further comprises a speaker module configured to convert the electrical signal into sound and output the sound into the user's oral cavity.

[0183] In some examples, the voice generation system is breath-driven and generates sound in response to vibrations of respiratory pressure (and / or airflow) in the oral cavity at a neck stoma. In some examples, the voice generation system further includes a voice enhancement module (such as a processing system 150) between the conversion module and the speaker module. In some examples, the voice enhancement module is a hardware or software module that improves the quality and volume of the sound.

[0184] In some examples, the voice generation system further generates or receives an airflow component in addition to voice, and outputs the voice and airflow components into the user's oral cavity. In some examples, the housing is configured to attach to the user's neck such that the first opening 112 connects to a neck stoma.

[0185] In some examples, the voice generation system further includes a pressure / airflow sensing module that attaches to the user's neck and is configured to detect air pressure or airflow in the user's neck stoma. In some examples, the voice generation system further includes a pressure / airflow sensing module configured to detect air pressure or airflow in the mouth. In some examples, the pressure / airflow sensing module samples the respiratory signal at 1 kHz.

[0186] In some embodiments, it is desirable to reduce (sound) interference (e.g., acoustic feedback or echo) received by the conversion module. Such interference may result from unwanted feedback from the sound output from the speaker module being picked up by the conversion module. Furthermore, such interference may also be caused by unwanted feedback from the speech signal being picked up by the conversion module, thereby causing the sound output from the speaker module to be modulated by the user's mouth or lip movements. Such interference is particularly problematic when it weakens the electrical signal of the conversion module converted from the vibration of the movable member and is amplified to result in unwanted positive feedback. In these situations, the processing system 150 may be configured to reduce the sound interference received by the conversion module 130. Being able to configure the processing system 150 to reduce the sound interference received by the conversion module is advantageous for ultimately improving the quality of the sound output from the speaker module 140.

[0187] In this specification, interference noise reduction may be considered to include, but not limited to, minimizing, canceling, suppressing, avoiding, or eliminating interference noise.

[0188] In this specification, “sound interference” is interchangeable with “interfering sound” and is intended to include, but is not limited to, acoustic feedback, echoes, ambient noise, and unwanted feedback sounds resulting from speech signals generated when sounds originating from the speaker module 140 or output from the speaker module into the user’s oral cavity stimulate the user’s vocal tract.

[0189] For example, as shown in Figure 1B, it is desirable to reduce interference noise 160 (which is interchangeable with “sound interference” in this specification). Interference noise 160 may include undesirable feedback noise originating from the speaker module 140, such as speaker module noise 170a and / or speech 170b. In context, speech 170b is the sound produced when the sound output from the speaker module to the user’s oral cavity stimulates the user’s vocal tract (oral cavity). However, interference noise 160 may include other forms of undesirable interference noise from other sound sources. Interference noise 160 may also include (acoustic) echo or acoustic feedback.

[0190] In some embodiments, the processing system 150 is configured to reduce the sound interference 160 based on at least one electrical signal received from a conversion module 130 configured to convert vibrations of a movable member, and / or at least one interference converter (e.g., interference conversion module 180 and / or interference conversion module 190) for receiving sound interference.

[0191] In some embodiments, the processing system 150 may receive an electrical input signal from the conversion module 130 to reduce sound interference. In such embodiments, the interfering sound 160 picked up by the conversion module 130 is used by the processing system 150 to reduce the interfering sound.

[0192] Additionally or alternatively, the processing system 150 may receive electrical input signals from at least one interference transformer module (different from the transformer module 130) to identify and reduce sound interference. In such embodiments, the interference sound 160 picked up by at least one interference transformer module is used by the processing system 150 to reduce the interference sound. For example, a speaker module sound 170a picked up by a first interference transformer module 180 may be used by the processing system 150 to reduce the interference sound. In another example, the processing system 150 may be used to reduce the interference sound when the first interference transformer module 180 picks up the speaker module sound 170a and the second interference transformer module 190 picks up speech 170b. Further interference transformer modules may be used as needed.

[0193] In some examples, the interference conversion modules (first and / or second conversion modules 180, 190, etc.) comprise one or more microphones or microphone arrays. In some examples, the interference conversion modules (first and / or second conversion modules 180, 190, etc.) comprise one or more of a piezoelectric transducer, a magnetic pickup transducer, an accelerometer, or any other sound, tone, or vibration sensor, which have the advantage of identifying or reducing acoustic interference from an external sound source.

[0194] In some embodiments, the processing system 150 is configured to reduce sound interference by performing feedforward suppression (interchangeable with "forward suppression"). In such embodiments, the feedforward suppression may include one or more uncorrelated operations (e.g., gain reduction, frequency-selective gain reduction, frequency-selective noise reduction, notch filtering, phase modulation, and frequency shifting). In some embodiments, the processing system 150 is configured to reduce sound interference by performing adaptive feedback cancellation. In such embodiments, the adaptive feedback cancellation may include adaptive filtering. Optionally, the adaptive filtering is performed using an FIR filter. Optionally, the adaptive filtering is performed using a (real-time) normalized least squares mean (NLMS) algorithm. In some embodiments, the processing system 150 is configured to reduce sound interference by performing residual feedback suppression.

[0195] In some embodiments, the processing system may include an AI algorithm trained to remove acoustic interference from the transducer signal. This may be performed with or without further interference transducers.

[0196] In some embodiments, a physical connection (e.g., an air connection) may be provided between the movable member, the user's mouth, and the user's stoma (so that they are all connected to each other). This could lead to saliva leaking from the mouth into the stoma (and possibly into the movable member, such as a membrane tube).

[0197] To avoid this, in such embodiments, the design may include a humidity sensor that detects the presence of saliva leakage or water near the stoma. If the presence of such saliva is detected, the voice generation system stops generating voices to clean the system and flashes an indicator light to warn the user of the presence of saliva.

[0198] In some embodiments, the voice generation system is configured to be used hands-free.

[0199] Up to this point, a detailed description of a general embodiment of the voice generation system has been given with reference to Figures 1A and 1B. Now, an example of an embodiment of the voice generation system will be described with reference to Figures 2 to 18.

[0200] Referring to Figure 2, another example of a voice generation system 200 is shown.

[0201] The voice generation system 200 comprises a housing 210 having a first opening 212 and a second opening 214. The housing 210 defines an air passage between the first opening 212 and the second opening 214. The first opening 212 is configured to communicate with the user 220's neck stoma. In some examples, the housing 210 is configured to adhere to the user 220's neck so that the first opening 212 connects to the neck stoma. The second opening 214 is configured to communicate with the user 220's oral cavity.

[0202] Therefore, the housing 210 is configured to adhere to the user 220's neck such that the first opening 212 connects to or contacts the neck stoma.

[0203] The voice generation system 200 further includes a tube 230 having a first open end 232 connected to the housing 210 around a second opening, and a second open end 234 of the tube configured to be inserted into the mouth of the user 220. In some examples, the tube 230 is a disposable tube.

[0204] The tube 230 is hollow and defines a second air passage (for both voice and airflow) between its first and second ends 232 and 234. In some examples, the tube 230 defines separate voice and airflow passages. The air passage of the tube 230 communicates with the air passage of the housing 210. Thus, the housing 210 and the tube 230 together define an air passage between the first opening 212 of the housing 210 and the second end 234 of the tube 230. During use, the air expelled from the neck stoma during exhalation travels along the air passage defined by the housing 210 and the tube 230 and is delivered into the user's mouth.

[0205] The tube 230 may be equipped with one or more filters or bends (e.g., a membrane air filter or a U-tube) to prevent or limit oral food residue, saliva, or other fluids that enter the second end 234 from reaching the first opening 212 of the housing 210 and thus the trachea of ​​the user 220.

[0206] The air pressure at the first opening 212 can correspond to the air pressure inside the neck stoma, while the air pressure at the second opening can correspond to the air pressure inside the user's 220 mouth.

[0207] The housing 210 and the tube 230 together define an air passage that extends between the neck stoma and the user 220's oral cavity. Thus, the respiratory airflow enters the first opening 212 of the housing 210, travels along the air passage of the housing 210, and then through the tube. 230 It is configured to travel along the air passage and be output to the mouth, or oral cavity, of the user 220.

[0208] The voice generation system 200 further includes a movable member 240 located within the housing and configured to vibrate in response to the flow of breathing air through the air passage of the housing 210.

[0209] The sound generation system 200 further includes a conversion module configured to convert the vibrations of the movable member 240 into electrical signals.

[0210] In some examples, the speech generation system 200 further includes a processing system with speech enhancement software / hardware to improve the quality or volume of the speech generated by the speech generation system 200. Artificial intelligence (AI) algorithms may be used in the processing system to improve the quality of the speech or even learn to imitate the user's natural voice.

[0211] The voice generation system 200 further includes a speaker module 250 located within the housing 210. The speaker module 250 is configured to convert electrical signals representing speech (received from the conversion module or the voice enhancement module) into sound and output the sound to the mouth of the user 220. The speaker module 250 is configured to output the sound through a second opening in the housing 210. In some examples, the speaker module uses a tube to transmit sound into the mouth of the user 230. 230 It is directed to output sound.

[0212] Therefore, in some examples, the voice generation system 200 combines the naturally occurring air pressure within the neck stoma and the naturally occurring air pressure within the user's oral cavity with the movable member 240. In some examples, the voice generation system 200 combines the natural airflow from breathing with the movable member 240. The resulting airflow or air pressure gradient experienced by the movable member 240 causes the movable member 240 to vibrate.

[0213] Figures 3 to 8 show another example of a voice generation system 300.

[0214] The voice generation system 300 comprises a housing 310 having a first opening and a second opening. The housing 310 defines an air passage between the first opening and the second opening. The first opening is a vent or air intake configured to allow air into the housing 310 to supply an air pump located inside the housing 310, while the second opening is configured to communicate with the mouth of the user 320.

[0215] The housing 310 may be configured to be fixed to the auricle, i.e., the protruding outer part of the ear, of the user 320, and is positioned between the user 320's auricle and head. In some examples, the housing 310 has an auricle shape or a spiral shape to facilitate fixation of the housing 310 to the person's auricle. In other examples, the housing 310 is a handheld housing.

[0216] The voice generation system 300 further includes a tube 330 having a first open end 332 connected to the housing 310 around a second opening, and a second open end 334 configured to be inserted into the oral cavity of the user 320. When the housing 310 is fixed to the auricle, the tube 330 is configured to bend around the outside of the user 320's cheek and reach the user 320's mouth.

[0217] The tube 330 is hollow and defines a second air passage (for both sound and airflow) between its first and second ends 332 and 334. The air passage of the tube 330 communicates with the air passage of the housing 310. Thus, the housing 310 and the tube 330 together define an air passage between the first opening of the housing 310 and the second end (or mouthpiece end) of the tube 330. In some examples, as shown in Figure 6, the tube 330 defines a voice passage 336 configured to transmit sound and an airflow passage 338 configured to transmit airflow. The voice passage 336 and the airflow passage 338 may be separated from each other.

[0218] The voice generation system 300 further includes an air pump 350 located within the housing 310. The air pump 350 is configured to generate an airflow that moves along an air passage from a first opening to a second opening.

[0219] The sound generation system 300 further includes a movable member 340, shown in Figure 7, which is located inside the housing and is configured to vibrate the airflow in the air passage generated by the air pump 350 in response to the air flowing through the air passage of the housing 310.

[0220] The sound generation system 300 further includes a conversion module 342 configured to convert the vibrations of the movable member 340 into electrical signals. As shown in Figure 7, the conversion module 342 may be located within the housing 310 and operably coupled to the movable member 340.

[0221] The voice generation system 300 may further include a processing system, which may include voice enhancement software / hardware modules, to improve the quality or volume of the voice generated by the movable member 340 and picked up via the conversion module 342. The processing system may use artificial intelligence (AI) to improve the quality of the voice or to learn to imitate the user's natural voice.

[0222] The voice generation system 300 further includes a speaker module 360 ​​located within the housing 310. The speaker module 360 ​​is configured to convert electrical signals representing speech (received from the conversion module 342 or the voice enhancement module) into sound and output the sound to the mouth of the user 320. The speaker module 360 ​​is configured to output sound through a second opening in the housing 310. As shown in Figure 7, the speaker module 360 ​​may include a speaker array distributed along the elongated tubular portion of the housing 310 near the second opening to better direct the sound into the tube 330.

[0223] The voice generation system 300 further includes a pressure sensing module 370 configured to be attached to the user 320's neck. The pressure sensing module 370 is configured to detect air pressure (and / or airflow) in the user 320's neck stoma. The pressure sensing module 370 is mounted to cover the stoma, or opening 372, located at the front of the user 320's neck.

[0224] The voice generation system 300 further includes a controller located within the housing 310. The controller is configured to control the air pump 350 based on air pressure / airflow detected in the neck stoma in real time. The controller is operably coupled to the air pump 350 to activate or operate the air pump 350. In some examples, the controller is located within the housing 310. In other examples, the controller forms part of the processing system.

[0225] The controller may be configured to receive or acquire measured, read, or displayed values ​​of air pressure from the pressure sensing module 370. In some examples, the pressure sensing module samples a breathing signal at 1 kHz and transmits it to the controller. In some examples, the controller communicates wirelessly with the pressure sensing module 370 using a Bluetooth® wireless link 380 and connectionless UDP protocol, etc. In some examples, to enable wireless communication between the controller and the pressure sensing module 370, the pressure sensing module 370 includes a wireless transmitter (such as a Bluetooth transmitter) and the controller includes a wireless receiver (such as a Bluetooth receiver).

[0226] The controller may control the air pump 350 such that the pump output is proportional to, otherwise dependent on, or a function of, the magnitude of the air pressure detected by the pressure sensing module 370. In some examples, the air pump 350 is configured to generate air pressure and / or airflow in the air passage corresponding to the air pressure or airflow in the neck stoma measured by the pressure sensing module 370. In some examples, the controller has a delay of less than 5 milliseconds to track the air pressure / airflow in the neck stoma in order to maintain real-time performance.

[0227] The controller may control the air pump 350 to generate airflows with different characteristics (e.g., airflow volume, airflow rate) based on the detected breathing air pressure. Furthermore, the characteristics of the airflow may affect the characteristics of the sound (e.g., pitch, frequency spectrum, or volume) produced by the movable member 340 when it vibrates. Thus, the sound produced by the movable member 340 depends on the breathing air pressure of the user 320 and the physical shape and materials used in the movable member 340.

[0228] Therefore, the voice generation system 300 monitors the airflow of the user 320's breathing in the neck stoma and reproduces the breathing airflow (in terms of air pressure and / or airflow) by sending wireless commands to the air pump 350 located inside the housing 310. The air pump 350 can therefore correspond to, be similar to, or otherwise generate the relevant air pressure and / or airflow of the user 320's breathing.

[0229] Thus, the airflow in the air passage in the housing 310 may correspond to the air pressure generated by the air pump 350, which in some examples may represent the air pressure in the neck stoma and the air pressure in the second opening that may correspond to the air pressure in the user's 320 oral cavity. Therefore, the movable member is affected on one side by the air pressure in the neck stoma artificially generated using the air pump 350, and on the other side by the air pressure in the oral cavity naturally generated by the user's 320. The resulting airflow or air pressure gradient experienced by the movable member 340 causes the movable member 240 to vibrate and generate sound.

[0230] Furthermore, the airflow generated by the air pump 350 is configured to travel along the air passages of the housing 310 and the air passages of the tube 330, and to be output to the mouth, or oral cavity, of the user 320 to provide an airflow that generates consonants when uttering.

[0231] In some examples, as shown in Figure 8, airflow and sound are generated within two distinct parts, or compartments, of the housing 310. The housing 310 comprises a first compartment 312 for generating airflow and a second compartment for synthesizing sound. Compartment 312 houses a movable member 340, a conversion module, and an air pump 350, while compartment 314 houses a speaker module 360 ​​and electronics for the operation of the sound generation system 300 (such as a battery 390, and a sound enhancement and / or conversion unit 392 (hardware and software, etc.) with a processing system for processing electrical signals that adjust the quality of the sound generated by the speaker module 360).

[0232] Sections 312 and 314 may be coupled or connected to communicate with pipe 330. In this way, airflow and sound may be combined so that both travel along the same air passage into the oral cavity of user 320.

[0233] Advantageously, the voice generation system 300 does not require a physical pathway between the user's mouth and breathing pathway (stoma).

[0234] Referring to Figure 9, an example of a movable member 340 suitable for use in the voice generation system 300 or any other voice generation system is shown. The movable member 340 comprises a single membrane 344. The membrane 344 is circular. Multiple holes 346 are arranged on the outer edge of the membrane 344 to allow the membrane 344 to be mounted in the internal space of the housing of the voice generation system. In other examples, the membrane 344 may have any other shape or may represent a simple or complex mechanical model of the vocal cords, such as a synthetic multilayer silicone model of the vocal cords (e.g., as described in "Synthetic, multi-layer, self-oscillating vocal fold model fabrication" by Murray, Preston R., and Scott L. Thomson, JoVE (Journal of Visualized Experiments) 58 (2011):e3498). The membrane 344 may be formed of silicone, natural rubber, or any other flexible material, and may comprise a single layer or multiple layers (e.g., filled with a gel-type material). The movable member 340 may comprise one or more vibrating elements and may have a non-uniform 3D pattern to generate irregular vibrations so that the resulting sound is not monotonous but sounds more natural.

[0235] Referring to Figures 10 and 11, an adhesive is shown for attaching an example pressure sensing module 370 to the user's neck. The pressure sensing module 370 may be suitable for use with a voice generation system 300 or any other voice generation system. The pressure sensing module 370 is configured to be attached to or bonded to an adhesive strip 374 so as to cover a hole in the adhesive strip 374. The strip 374 has an adhesive surface configured to adhere to the skin of user 320 for attaching the pressure sensing module 370 to the neck or chest of user 320 so as to cover a neck stoma. The strip 374 may hold a heat and moisture exchanger (HME) cap 376 over the stoma of a pharyngectomy patient to protect the airway of the pharyngectomy patient from external contamination. The pressure sensing module 370 may be located inside the HME cap 376.

[0236] Referring to Figures 12 and 13, another example of a voice generation system 400 is shown.

[0237] The voice generation system 400 comprises a first housing 410 having a first opening 412 and a second opening 414. The housing 410 defines an air passage between the first opening 412 and the second opening 414. The first opening 412 is configured to communicate with the user 420's neck stoma, while the second opening 414 may communicate with the external environment of the housing 410 and the user 420 or with a silencer (described later). Thus, the airflow to the first opening 412 corresponds to the airflow to the user 420's neck stoma due to breathing.

[0238] In some examples, as shown in Figure 12, the housing 410 is configured to attach directly to the front of the user 420's neck such that the first opening connects to or is close to the neck stoma. In other examples, as shown in Figure 13, the housing 410 is attached to or integrated with a neck harness 416 having an arched or U-shaped frame configured to rest on the user 420's shoulders and around the back of the neck.

[0239] The sound generation system 400 further includes a movable member 430 positioned between the first opening 412 and the second opening 414 within the housing 410 and configured to vibrate in response to air flowing through the air passage of the housing 410.

[0240] The sound generation system 400 further includes a conversion module configured to convert the vibrations of the movable member 430 into electrical signals. In some examples, the conversion module is located within the housing 410.

[0241] In some embodiments of the sound generation system 400, in which the second opening 414 may be connected to the outside air, the sound of the moving member becomes monotonous because the air pressure in or near the second opening 414 is not connected to the oral cavity or oral canal. To solve this problem, the oral pressure airflow may be monitored to improve and adjust the quality of the sound signal. Thus, the sound generation system 400 is more strongly influenced by fluctuations in pressure in the mouth in order to maintain a non-monotonous sound. That is, in these embodiments, the sound generation system produces a sound output based on monitored air pressure.

[0242] The voice generation system 400 further includes a regulator, or airflow control element 440, which is located within the housing 410 and configured to control the airflow in the air passage.

[0243] In some examples, the regulator 440 is a first air pump configured to generate air pressure and / or airflow in or near the second opening 414 within the air passage. The air pressure generated by the first air pump may correspond to the air pressure in the user's mouth or oral canal. The difference between the stoma air pressure at the first opening 412 and the pressure generated by the first air pump at the second opening 414 creates an airflow that moves along the air passage from the first opening 412 to the second opening 414, causing the movable member 430 to vibrate.

[0244] In other examples, the regulator 440 is an air valve configured to control the airflow through the second opening 414. The air valve may allow for the adjustment of the airflow entering and leaving the air passage through the second opening 414. In some examples, the air valve is an electromechanically actuated valve (such as a miniature solenoid valve) which may provide a controlled inlet and outlet for air to alter (e.g., open and close) the air passage in a quiet mode with low latency (e.g., millisecond delay) and linear and low audible noise (e.g., 20 dB). The air valve may increase or decrease the airflow by being actuated between an open and closed state. The air valve may have one or more partially open or partially closed states to provide continuous control or fine adjustment of the fluid flow between the open and closed states. Thus, by acting the air valve, the pressure and / or airflow in the air passage near the second opening 414 may be controlled. The airflow and / or pressure of the breath produced by the user 420 through the neck stoma causes the movable member 430 to vibrate when regulated by the air valve. In other examples, the regulator 440 is a suction device. In other examples, the regulator 440 is an actuator. The actuator may be a miniactuator or a microactuator.

[0245] The voice generation system 400 further comprises a second housing 450 which may be configured to be fixed to the auricle of the user 420. The housing 450 is separate from the housing 410. The housing 450 comprises a first opening which is a vent or air intake configured to allow air into the housing 450 to supply an air pump housed inside (i.e., the second air pump of the voice generation system 400), and a second opening which communicates with the oral cavity of the user 420. The housing 450 defines an air passage between the first and second openings. In other examples, the housing 450 comprises a single opening which communicates with the oral cavity of the user 420.

[0246] The voice generation system 400 further includes a tube 460 having a first open end 462 connected to a second opening of the housing 450 and a second open end 464 configured to be inserted into the oral cavity of the user 420.

[0247] The voice generation system 400 further includes an air pump located within the housing 450. The air pump in the housing 450 may be referred to as the second air pump to distinguish it from the first air pump of the regulator 440. The second air pump is configured to generate an airflow into the user's oral cavity 420. In some examples, the airflow generated by the second air pump corresponds to or represents the airflow from the neck stoma due to breathing in the housing 410.

[0248] The voice generation system 400 further includes a flow detection module 418 located within the housing 410. In some examples, the flow detection module 418 is located within the air passage of the housing 410. The flow detection module 418 is configured to detect or measure the airflow in the neck stoma due to breathing or the airflow (such as airflow rate and / or volume) from the neck stoma into the air passage of the housing 410. In some examples, the flow detection module 418 includes a differential pressure sensor or two or more pressure sensors. In some examples, the flow measurement performed by the flow detection module 418 is used to operate a second air pump in the housing 450 to generate an airflow corresponding to the airflow from the neck stoma.

[0249] The voice generation system 400 further includes a speaker module located within the housing 450. The speaker module is configured to convert the electrical signal generated by the conversion module into sound and output the sound to the mouth of the user 420. The speaker module is configured to output the sound through a second opening in the housing 450. In some examples, the speaker module 450 is configured to receive the electrical signal (a signal representing sound) generated by the conversion module via a wired or wireless link between the housing 410 and the housing 450.

[0250] In some cases, the sound generated by the vibration of the movable member 430 may interfere with or interfere with the sound (representing speech) generated by the speaker module, and therefore it may be desirable to suppress it. Accordingly, in some cases, the speech generation system 400 further includes a silencer or other noise reduction element to reduce or suppress undesirable background noise generated by the vibration of the movable member 430. The silencer may be provided within the first housing 410 (e.g., in an air passage) or connected to the second opening 414.

[0251] The voice generation system 400 further includes a pressure sensing module 470 configured to detect the air pressure in the user's 420 oral cavity. In some examples, the pressure sensing module 470 is configured to measure a pressure signal in the oral cavity. In some examples, the pressure sensing module 470 is attached to the second end of the tube 460 so that it is positioned inside the user's 420 oral cavity during use. In some examples, the pressure sensing module 470 includes an intraoral pressure sensor.

[0252] The voice generation system 400 further includes a controller located within the housing 410. The controller is configured to control the regulator 440 based on the intraoral air pressure detected by the pressure sensing module 470. In some examples, the controller communicates wirelessly with the pressure sensing module 470, such as via a Bluetooth wireless link 480. In some examples, to enable wireless communication between the controller and the pressure sensing module 470, the pressure sensing module 470 includes a wireless transmitter (such as a Bluetooth transmitter), and the controller includes a wireless receiver (such as a Bluetooth receiver).

[0253] The voice generation system 400 further includes a second controller located within a second housing 450. The second controller is configured to control a second air pump based on the detected airflow from the neck stoma, measured by a flow sensing module 418. In some examples, the second controller communicates wirelessly with the flow sensing module 418, for example, via a second Bluetooth wireless link. In some examples, to enable wireless communication between the second controller and the flow sensing module 418, the flow sensing module 418 includes a wireless transmitter (such as a Bluetooth transmitter), and the second controller includes a wireless receiver (such as a Bluetooth receiver).

[0254] In some examples, if the regulator 440 is an air pump, the air pump is configured to generate air pressure in the air passage that corresponds to the air pressure inside the oral cavity measured by the pressure sensing module 470. In other examples, if the regulator 440 is an air valve, it is configured to control the airflow in the air passage based on the air pressure inside the oral cavity measured by the pressure sensing module 470 (the air pressure inside the oral cavity may have an inverse relationship with or inverse effect on the airflow in the valve, such that the airflow decreases as the air pressure inside the oral cavity increases).

[0255] Referring to Figure 14, another voice generation system 500 is shown.

[0256] The voice generation system 500 comprises a first housing 510 having a first opening 512 and a second opening 514. The housing 510 defines an air passage between the first opening 512 and the second opening 514. The first opening 512 is configured to communicate with the user 520's neck stoma, while the second opening 514 may communicate with the external environment of the housing 510 and the user 520 or with the (previously described) silencer. Therefore, the airflow to the first opening 512 corresponds to the airflow to the user 520's neck stoma.

[0257] In some examples, the housing 510 is configured to attach directly to the front of the user 520's neck such that the first opening connects to or is in close proximity to the neck stoma. In other examples, the housing 510 is attached to or integrated with a neck harness having an arched or U-shaped frame configured to rest on the user 520's shoulders and around the back of the neck (in an arrangement similar to that shown in Figure 13).

[0258] The sound generation system 500 further includes a movable member 530 positioned between the first opening 512 and the second opening 514 within the housing 510 and configured to vibrate in response to air flowing through the air passage of the housing 510.

[0259] The sound generation system 500 further includes a conversion module configured to convert the vibrations of the movable member 530 into electrical signals. In some examples, the conversion module is located within the housing 510.

[0260] In some embodiments of the sound generation system 500, in which the second opening 514 may be connected to the outside air, the sound of the moving member becomes monotonous because the air pressure in or near the second opening 514 is not connected to the oral cavity or oral canal. To solve this problem, the oral pressure airflow may be monitored to improve and adjust the quality of the sound signal. Thus, the sound generation system 500 is more strongly influenced by fluctuations in pressure in the mouth in order to maintain a non-monotonous sound. That is, in these embodiments, the sound generation system produces a sound output based on monitored air pressure.

[0261] The voice generation system 500 further includes a regulator, or airflow control element 540, which is located within the housing 510 and configured to control the airflow in the air passage.

[0262] In some examples, the regulator 540 is a first air pump configured to generate air pressure and / or airflow in or near the second opening 514 within the air passage. The air pressure generated by the first air pump may correspond to the air pressure in the user's mouth or oral canal. The difference between the stoma air pressure at the first opening 512 and the pressure generated by the first air pump at the second opening 514 creates an airflow that moves along the air passage from the first opening 512 to the second opening 514, causing the movable member 530 to vibrate.

[0263] In other examples, the regulator 540 is an air valve configured to control the airflow through the second opening 514. The air valve may allow for the adjustment of the airflow entering and leaving the air passage through the second opening 514. In some examples, the air valve is an electromechanically actuated valve (such as a miniature solenoid valve) which may provide a controlled inlet and outlet for air to alter (e.g., open and close) the air passage in a quiet mode with low latency (e.g., millisecond delay) and linear and low audible noise (e.g., 20 dB). The air valve may increase or decrease the airflow by being actuated between an open and closed state. The air valve may have one or more partially open or partially closed states to provide continuous control or fine adjustment of the fluid flow between the open and closed states. Thus, by acting the air valve, the pressure and / or airflow in the air passage near the second opening 514 may be controlled. The airflow and / or pressure of breathing, generated by the user 520 through the neck stoma and regulated by the air valve, causes the movable member 530 to vibrate. In other examples, the regulator 540 is a suction device. In other examples, the regulator 540 is an actuator. The actuator may be a miniactuator or a microactuator.

[0264] The voice generation system 500 further comprises a denture unit, mouthplate, or frame 550 configured to be fixed in the oral cavity of the user 520, for example, on the palate. The mouthplate 550 is separate from the housing 510. The mouthplate 550 comprises a frame that is open to or in communication with the oral cavity of the user 520.

[0265] The voice generation system 500 further includes an air pump 560 connected to a mouthplate 550. The air pump on the mouthplate 550 may be called a second air pump to distinguish it from the first air pump of the regulator 540. In some examples, the air pump 560 is a micro air pump. The air pump 560 is configured to generate an airflow into the oral cavity of the user 520. In some examples, the airflow generated by the second air pump 560 corresponds to or represents the airflow from the neck stoma due to breathing, as measured within the housing 510.

[0266] The voice generation system 500 further includes a flow detection module 518 located within the housing 510. In some examples, the flow detection module 518 is located within the air passage of the housing 510. The flow detection module 518 is configured to detect or measure the airflow in the neck stoma due to breathing or the airflow (such as airflow rate and / or volume) from the neck stoma into the air passage of the housing 510. In some examples, the flow detection module 518 includes a differential pressure sensor or two or more pressure sensors. In some examples, the flow measurement performed by the flow detection module 518 is used to operate a second air pump 560 to generate an airflow corresponding to the airflow from the neck stoma.

[0267] The voice generation system 500 further includes a speaker module 570 attached to the mouthplate 550. The speaker module 570 is configured to convert the electrical signals generated by the conversion module into sound and output the sound to the mouth of the user 520. In some examples, the speaker module 570 is configured to receive the electrical signals (signals representing sound) generated by the conversion module via a wired or wireless link between the housing 510 and the mouthplate 550.

[0268] The voice generation system 500 further includes a pressure sensing module 580 attached to the mouthplate 550 (denture source). The pressure sensing module 580 is configured to detect the air pressure in the user's 520 oral cavity. In some examples, the pressure sensing module 580 is configured to sample respiratory signals in the oral cavity. In some examples, the pressure sensing module 580 includes an oral pressure sensor.

[0269] The voice generation system 500 further includes a controller located within the housing 510. The controller is configured to control the regulator 540 based on the air pressure detected in the oral cavity. In some examples, the controller communicates wirelessly with the pressure sensing module 580, such as via a Bluetooth wireless link 590. In some examples, to enable wireless communication between the controller and the pressure sensing module 580, the pressure sensing module 580 includes a wireless transmitter (such as a Bluetooth transmitter), and the controller includes a wireless receiver (such as a Bluetooth receiver).

[0270] The voice generation system 500 may further include a second controller mounted on the mouth plate 550. The second controller is configured to control a second air pump 560 based on the detected airflow from the neck stoma measured by the flow detection module 518. In some examples, the second controller communicates wirelessly with the flow detection module 518, for example, by a second Bluetooth wireless link 592, as shown in Figure 15. In some examples, to enable wireless communication between the second controller and the flow detection module 518, the flow detection module 518 includes a wireless transmitter (such as a Bluetooth transmitter) and the second controller includes a wireless receiver (such as a Bluetooth receiver).

[0271] In some examples, if the regulator 540 is an air pump, the air pump is configured to generate air pressure in the air passage that corresponds to the oral cavity pressure measured by the pressure sensing module 580. In other examples, if the regulator 540 is an air valve, it is configured to control the airflow in the air passage based on the oral cavity air pressure measured by the pressure sensing module 580 (the oral cavity air pressure may have an inverse relationship with or inverse effect on the airflow of the valve, such that the airflow decreases as the oral cavity air pressure increases).

[0272] The voice generation systems 400 and 500 use a "push-pull" mechanism to cause vibration of a movable member. That is, the airflow exiting the neck stoma during exhalation "pushes" the movable member, while the pressure generated from the operation of a regulator (e.g., either the first pump or the air valve) "pushes" the movable member. With respect to air pressure, the movable member is influenced on one side by the air pressure in the neck stoma naturally generated during breathing, and on the other side by the air pressure in the oral cavity artificially simulated using the regulator. The resulting airflow or air pressure gradient experienced by the movable member causes it to vibrate and generate sound.

[0273] In some cases, implementing a regulator as an air valve can facilitate miniaturization and power consumption of the voice generation system 400 or 500, and can improve the user experience as the air valve may be quieter than an air pump during operation.

[0274] The voice generation systems 400 and 500 may further include a processing system (in the housing 410 or 510 behind the conversion module, or in the housing 450 or denture unit 550 in front of the speaker module) that includes voice enhancement software / hardware to improve the quality or volume of the voice generated by the voice generation system. In some examples, the processing system includes an AI voice conversion module.

[0275] Referring to FIGS. 17 to 18, an audio generation system 700 of another example is shown.

[0276] As shown in FIG. 17, the audio generation system 700 is similar or identical to the audio generation system 200, except that the audio generation system 700 is used to illustrate how some embodiments of an audio generation system can be configured to reduce acoustic interference.

[0277] Referring to FIG. 17, an air flow and sound are generated within two different portions or compartments of the housing 710. The housing 710 includes a first compartment 716 for generating an air flow and a second compartment 718 for synthesizing sound. The first compartment 716 houses a movable member 740 and a conversion module 742 and communicates with the user's neck stoma 752, while the second compartment 718 houses a speaker module 750 and electronics (such as a battery 788 and an audio enhancement and / or conversion unit 792 (including hardware and software, etc.) having a processing system 794) for the operation of the audio generation system 700. The processing system 794 may be used to process electrical signals generated by the conversion module 742. The processing system 794 may be used to process electrical signals for adjusting the quality of sound generated by the speaker module 750. Additionally or alternatively, the processing system 794 may be configured to reduce acoustic interference received at the conversion module 742, which will be described in detail below.

[0278] Sections 716 and 718 may be coupled or connected to communicate with the tube 730. In this way, airflow, vibration / sound of movable members, and speaker module sound may be combined to travel together along the same air passage into the oral cavity of the user 720. However, in some examples, it is desirable to reduce sound interference 760 received by the conversion module 742. Sound interference 760 may be caused by the output of speaker module sound 770a propagating from the second section 718 to the first section 716. Sound interference 760 may also be caused by phonation 770b propagating from the tube 730 into the first section 716. Phonation 770b is generated as a result of sound (from the tube 730) leaving the vocal tract and being regulated by the user's mouth or lips.

[0279] In such circumstances, it is desirable that the sound generation system 700 has a processing system 794 configured to reduce sound interference received by the conversion module. The processing system 794 may be configured to reduce sound interference 760 based on at least one electrical signal received from a conversion module 742 configured to convert vibrations of the movable member 740. Additionally or alternatively, the processing system 794 may be configured to reduce sound interference 760 based on at least one electrical signal received from at least one interference conversion module (which may be interference conversion module 780 and / or interference conversion module 790) for receiving sound interference 760. Further explanation is provided with reference to Figure 18.

[0280] In some embodiments, as shown in Figure 18, the stoma airflow drives a movable member 740 (in this example, a membrane) to generate sound e(t). e(t) is transmitted to the oral cavity via a tube (membrane tube 730b). A conversion module 742, in this example, is a microphone and is located near the membrane. The microphone picks up the membrane sound e(t) and sends it to a speaker module 750 (in this example, a loudspeaker). The loudspeaker amplifies the sound and sends e(t) to the oral cavity via a second tube (speaker tube 730a). The speaker tube and the membrane tube merge at some point to form an oral cavity tube 730, which sends the sound and airflow to the oral cavity.

[0281] The problem with these tubes converging toward the oral cavity is that the speaker sound can find a way back through tube 730 to reach membrane tube 730b through the oral cavity as acoustic feedback e'(t). This is particularly problematic when the speaker sound gets louder or when the membrane becomes quiet during the transition between voiced and voiceless sounds. If the patient tries to speak louder, the speaker gain amplifies e'(t) compared to e(t), causing the microphone to start picking up e'(t) instead of e(t), creating a positive feedback loop between the microphone and the speaker. The same problem exists during the voiced / voiceless transition of a voice-generating system. The membrane automatically generates sound in voiced sounds and becomes quiet in voiceless sounds (driven by pressure fluctuations in the patient's mouth and stoma). When the membrane is generating sound, the microphone, being close to the membrane, typically picks up the membrane sound e(t). As the membrane begins to quiet during the voiced-to-voiceless transition, the membrane sound e(t) weakens towards zero. However, the intraoral speaker sound e'(t) does not weaken at the same time. The oral cavity is a resonant cavity, which means it retains the speaker sound e'(t) inside for a longer period. Therefore, the membrane microphone begins to pick up e'(t), and an acoustic echo or feedback loop is formed during the voiced / voiceless transition.

[0282] Another potential sound interference issue in this system is the vocalization signal s(t) naturally produced by the mouth using the speech generation system. Embodiments of the speech generation system described herein relate to membrane sounds and speaker sounds e(t)+e'(t) for stimulating the mouth to produce a vocalization signal (s(t)) naturally produced by the user when the facial / lip muscles are moved. In a similar scenario, in addition to the speaker sound e'(t), the vocalization signal s(t) also finds a path to reach the membrane microphone as undesirable interference.

[0283] If sufficient acoustic feedback reduction (e.g., suppression) is not properly implemented in the system, the feedback / noise signal e'(t)+s(t) will weaken the membrane microphone signal, forming a positive feedback loop.

[0284] As explained above, there are two identifiable feedback paths within this system. The feedback reduction module identifies and measures these paths and removes them from the membrane microphone signal in real time. These feedback paths are as follows: 1. Speaker feedback sound (e'(t)). This can be measured by the first interference transformer module 780, which in this example is a microphone, referred to as "microphone 1," and is located in a speaker tube (microphone 1) near the mouth. 2. Speech signal s(t) feedback. This can be measured by the second interference transformer module 790, which in this example is an ambient microphone, referred to as "microphone 2," and is exposed to the environment, allowing it to monitor the airborne speech signal s(t) as it is transmitted outside the mouth.

[0285] The feedback reduction system monitors the speaker feedback signal (e'(t)) and the vocal interference signal s(t) in real time. Then, it eliminates unwanted feedback (noise) using one or both of the following two main approaches: 1) Forward suppression 2) Adaptive feedback cancellation or residual feedback reduction approach.

[0286] Forward suppression approaches aim to prevent feedback from occurring in the initial stages. Adaptive feedback cancellation or residual feedback reduction tend to reduce feedback after it has been present.

[0287] Typical examples of forward suppression applicable to this system include automatic gain control such as frequency-selective gain control (which automatically limits speaker gain to specific frequencies related to the feedback path to avoid feedback), speaker and microphone phase modulation, or frequency shifting to make it easier for a membrane microphone to distinguish between e(t) and e'(t).

[0288] These examples of forward suppression are not very useful for louder speech or may reduce the speech signal. Therefore, in some embodiments, adaptive feedback cancellation or residual feedback suppression methods may be beneficial. The main method of adaptive feedback cancellation is adaptive filtering, or utilizing an AI algorithm that learns to distinguish between the membrane sound e(t) and the speech signal s(t).

[0289] In some embodiments, the Normalized Least Squares Mean (NLMS) algorithm is used for adaptive feedback cancellation fast enough for real-time applications. These adaptive feedback methods use microphone signals (microphone 1 780 and microphone 2 790) to recursively approximate the feedback path and remove the acoustic feedback signal from the membrane microphone signal. They work by minimizing the error signal between e(t) + echo and e(t), where the echo is (e'(t) + s(t) provided by the microphones (microphone 1 780 and microphone 2 790).

[0290] Furthermore, microphones 1 and the "membrane microphone" are used in some embodiments to further assist this interference reduction approach by estimating the transfer function of tube 730, which further modifies the speaker feedback e'(t) as the tube returns to the membrane microphone.

[0291] Feedback reduction in speech generation systems is further enhanced by voiced / voiceless determination. This helps eliminate feedback during voiced / voiceless transitions when the speech generation system turns off the microphone in real time to stop voiceless sounds. The use of voiced / voiceless determination for feedback reduction is described in more detail above.

[0292] Referring to Figure 16, a method 600 for generating sound is shown. The method 600 comprises a step 610 of providing a movable member that communicates with (interacts with) an airflow corresponding to the airflow of the user's breathing. The method 600 then comprises a step 620 of detecting vibrations of the movable member in response to the airflow. The method 600 then comprises a step 630 of generating sound derived from the detected vibrations using one or more electric speakers. The method 600 then comprises a step 640 of supplying airflow and sound into the user's oral cavity.

[0293] In some cases, the airflow supplied in process 640 is the same airflow that communicates with (interacts with) the movable member (e.g., the breathing airflow). In other cases, the airflow supplied in process 640 is a different airflow from the one that communicates with (interacts with) the movable member.

[0294] In some examples, the movable member is provided within an air passage that is subjected to a first air pressure at a first end of the air passage and a second air pressure at a second end of the air passage. In some examples, the first air pressure corresponds to the air pressure in the user's neck stoma, where the second air pressure corresponds to the air pressure in the user's oral cavity.

[0295] In some examples, method 600 further comprises detecting the air pressure and / or air flow (i.e., breathing air flow) at the user's neck stoma, and generating a first air pressure and air flow (i.e., breathing air flow) at the neck stoma using an air pump. In some examples, method 600 further comprises detecting the air pressure within the user's oral cavity, and generating a second air pressure using an air flow control element or regulator. In some examples, the air flow control element is an air pump. In some examples, the air flow control element is an air valve.

[0296] In some examples, method 600 further comprises guiding or directing the user's breathing air flow within an air passage, where the air flow is the directed breathing air flow.

[0297] In some examples, method 600 further comprises converting the vibration of a movable member into an electrical signal using a transducer. Method 600 further comprises processing the electrical signal (representing speech) before supplying the processed electrical signal to one or more electrical speakers for generating sound. The step of processing the electrical signal may include improving or enhancing the volume (e.g., increasing the loudness of the sound) or quality of the speech encoded or represented by the electrical signal. The processing step may include AI-based speech conversion software / hardware to improve the acoustic characteristics or naturalness of the speech.

[0298] In some examples, method 600 further comprises reducing the sound interference generated by one or more electrical speakers.

[0299] Optional embodiments may be considered to broadly include parts, elements, processes, and / or features mentioned or shown herein, individually or in any combination of two or more of these parts, elements, processes, and / or features, and where a particular integer having a well-known equivalent in the art to which the invention relates is mentioned, such well-known equivalent is considered to be incorporated herein as if it were described individually.

[0300] While preferred embodiments have been described in detail, it should be understood that many variations, modifications, substitutions, or alterations will become apparent to those skilled in the art without departing from the scope of the present invention.

[0301] Throughout this specification and the following claims, unless the context requires otherwise, the term “comprise” and its variations (such as “comprises” or “comprising”) are understood to include the integer or process or group of integers or processes mentioned, but not to exclude any other integer or process or group of integers or processes. Disclosed features 1. It is a voice generation system, A housing having a first opening and a second opening, wherein the housing defines an air passage between the first opening and the second opening, A movable member is disposed within the housing and configured to vibrate in accordance with the flow of air in the air passage, A conversion module configured to convert the vibration of the movable member into an electrical signal, A speaker module configured to convert the aforementioned electrical signal into sound and output the sound into the user's oral cavity, A voice generation system equipped with the following features. 2. A sound generation system as described in Clause 1, wherein the flow of air through the air passage is due to the difference in air pressure between the first opening and the second opening. 3. A voice generation system according to Clause 1 or 2, wherein the first opening is configured to communicate with the user's neck stoma. 4. A voice generation system as described in Clause 3, wherein the housing is configured to be attached to the user's neck or chest such that the first opening connects to the neck stoma. 5. A voice generation system according to any one of clauses 1 to 4, wherein the second opening is configured to communicate with the user's oral cavity. 6. A voice generation system according to any one of clauses 1 to 5, further comprising a tube having a first end connected to the housing around the second opening, the second end of the tube being configured to be inserted into the user's oral cavity. 7. A sound generation system according to Clause 6, wherein both the housing and the tube define an air passage between the first opening of the housing and the second end of the tube. 8. A sound generating system according to any one of clauses 1 to 7, wherein the speaker module is located within the housing and the speaker module is configured to output sound through the second opening. 9. A voice generation system as described in Clause 1, wherein the housing is configured to be connected to an airflow source, and the airflow source is configured to generate an airflow and output the airflow into the user's oral cavity. 10. A voice generation system as described in Clause 9, wherein the airflow source is the user's neck stoma, and the airflow is the airflow of respiration output from the user through the neck stoma. 11. A voice generation system as described in Clause 9, wherein the air source is an air pump. 12. The voice-generating system described in Clause 1, further, A pressure sensing module, which is attached to the user's neck and configured to detect the air pressure in the user's neck stoma, An air pump disposed within the housing and configured to generate an airflow that moves along the air passage from the first opening to the second opening, A controller configured to control the air pump based on the detected air pressure in the neck stoma, A voice generation system equipped with the following features. 13. A sound generation system as described in Clause 12, wherein the flow of air through the air passage is due to the difference between the air pressure in the air passage and the air pressure at the second opening. 14. A voice generation system according to clause 12 or 13, wherein the second opening is configured to communicate with the user's oral cavity. 15. A voice generation system according to any one of clauses 12 to 14, further comprising a tube having a first end connected to the housing around the second opening, the second end of the tube being configured to be inserted into the user's oral cavity. 16. A sound generating system as described in Clause 15, wherein both the housing and the tube define an air passage between the first opening of the housing and the second end of the tube. 17. A sound generating system according to any one of clauses 12 to 16, wherein the speaker module is located within the housing and is configured to output sound through the second opening. 18. A voice generation system according to any one of clauses 12 to 17, wherein the air pump is configured to generate air pressure in the air passage corresponding to the detected air pressure in the neck stoma. 19. A voice generation system as described in any one of clauses 12 to 18, wherein the housing is configured to be fixed to the user's auricle. The present invention can also be realized in the following embodiments, for example. Application Example 1: It is a voice generation system, A housing having a first opening and a second opening, wherein the housing defines an air passage between the first opening and the second opening, A movable member is disposed within the housing and configured to vibrate in accordance with the flow of air in the air passage, A conversion module configured to convert the vibration of the movable member into an electrical signal, A speaker module configured to convert the aforementioned electrical signal into sound and output sound into the user's oral cavity, Equipped with, The voice generation system is configured such that, during use, the air pressure at or near the second opening corresponds to the air pressure inside the user's mouth. Application example 2: A sound generation system according to claim 1, wherein the flow of air through the air passage is due to the difference in air pressure between the first opening and the second opening. Application Example 3: A voice generation system according to claim 1 or 2, wherein the air pressure at or near the first opening corresponds to the air pressure at the user's neck. Application Example 4: A voice generation system according to claim 3, wherein the air pressure at or near the first opening corresponds to the air pressure in the user's neck stoma. Application Example 5: A voice generation system according to any one of the above claims, wherein the first opening is configured to communicate with the user's neck stoma such that the air pressure in the first opening corresponds to the air pressure in the user's neck stoma. Application example 6: A voice generation system according to any one of the above claims, wherein the second opening is configured to communicate with the user's oral cavity such that the air pressure in the second opening corresponds to the air pressure in the user's oral cavity. Application example 7: A voice generation system according to any one of the above claims, further comprising a tube having a first end connected to the housing around the second opening, the second end of the tube being configured to be inserted into the user's oral cavity. Application Example 8: A sound generation system according to claim 7, wherein both the housing and the tube define an air passage between the first opening of the housing and the second end of the tube. Application example 9: A sound generation system according to any one of the above claims, wherein the speaker module is located within the housing and the speaker module is configured to output sound through the second opening. Application Example 10: A voice generation system according to any one of the above claims, wherein the housing is configured to be connected to an airflow source, and the airflow source is configured to generate an airflow and output the airflow into the user's oral cavity. Application Example 11: A voice generation system according to claim 10, wherein the airflow source is the user's neck stoma, and the airflow is the airflow of breathing output from the user through the neck stoma. Application Example 12: A voice generation system according to claim 10, wherein the air source is an air pump. Application Example 13: A voice generation system according to claim 1, further, A pressure sensing module, which is attached to the user's neck and configured to detect the air pressure in the user's neck stoma, An air pump disposed within the housing and configured to generate an airflow that moves along the air passage from the first opening to the second opening, A controller configured to control the air pump based on the detected air pressure in the neck stoma, A voice generation system equipped with the following features. Application Example 14: A sound generation system according to claim 13, wherein the flow of air through the air passage is due to the difference between the air pressure in the air passage and the air pressure at the second opening. Application Example 15: A voice generation system according to claim 13 or 14, wherein the second opening is configured to communicate with the user's oral cavity. Application Example 16: A voice generation system according to any one of claims 13 to 15, further comprising a tube having a first end connected to the housing around a second opening, the second end of the tube being configured to be inserted into the user's oral cavity. Application Example 17: A sound generation system according to claim 16, wherein both the housing and the tube define an air passage between the first opening of the housing and the second end of the tube. Application Example 18: A sound generation system according to any one of claims 13 to 17, wherein the speaker module is located within the housing and is configured to output sound through the second opening. Application Example 19: A voice generation system according to any one of claims 13 to 18, wherein the air pump is configured to generate an air pressure in the air passage corresponding to the detected air pressure in the neck stoma. Application Example 20: A voice generation system according to any one of claims 13 to 19, wherein the housing is configured to be fixed to the user's auricle. Application Example 21: A voice generation system according to any one of the above claims, wherein the voice generation system is configured to be used hands-free. Application Example 22: It is a voice generation system, A housing having a first opening and a second opening, wherein the housing defines an air passage between the first opening and the second opening, and the first opening is configured to communicate with the user's neck stoma, A movable member is disposed within the housing and configured to vibrate in accordance with the flow of air in the air passage, A regulator, which is disposed within the housing and configured to control the airflow in the air passage, A conversion module configured to convert the vibration of the movable member into an electrical signal, A speaker module configured to convert the electrical signal generated by the conversion module into sound and output sound into the user's oral cavity, An air pump configured to generate an airflow into the user's oral cavity, A pressure sensing module configured to detect the air pressure inside the user's oral cavity, A controller configured to control the regulator based on the detected air pressure in the oral cavity, A voice generation system equipped with the following features. Application Example 23: A voice generation system according to claim 22, wherein the housing is configured to be attached to the user's neck such that the first opening is connected to the neck stoma. Application Example 24: A voice generation system according to claim 22, wherein the housing is attached to a neck harness configured to fit around the back of the user's neck. Application Example 25: A voice generation system according to any one of claims 22 to 24, further comprising: A flow detection module is disposed within the housing and configured to detect the airflow in the air passage. A second controller is configured to control the air pump based on the detected airflow within the air passage, Equipped with, A voice generation system wherein the air pump is configured to generate an airflow to the oral cavity corresponding to the detected airflow in the air passage. Application Example 26: A sound generation system according to any one of claims 22 to 25, wherein the air pump and the speaker module are housed in a second housing configured to be fixed to the auricle of the user, the second housing having an opening configured to communicate with the user's oral cavity, and the speaker module is configured to output sound through the opening of the second housing. Application Example 27: A voice generation system according to claim 26, further comprising a tube having a first end connected to the opening of the second housing, the second end of the tube being configured to be inserted into the oral cavity of the user, and the pressure sensing module being attached to the second end of the tube. Application Example 28: A voice generation system according to any one of claims 22 to 25, wherein the air pump, the speaker module, and the pressure sensing module are attached to a denture unit configured to be fixed in the user's oral cavity. Application Example 29: A voice generation system according to any one of claims 22 to 28, wherein the regulator is a further air pump configured to generate in the air passage an air pressure corresponding to the detected air pressure in the oral cavity. Application Example 30: A voice generation system according to any one of claims 22 to 28, wherein the regulator is an air valve configured to control the airflow through the second opening. Application Example 31: A voice generation system according to any one of the above claims, wherein the pressure detection module communicates wirelessly with the controller, and the conversion module communicates wirelessly with the speaker module. Application Example 32: A voice generation system according to any one of the above claims, further comprising a processing system. Application Example 33: A voice generation system according to claim 32, wherein the processing system is configured to reduce sound interference received by the conversion module. Application Example 34: A voice generation system according to claim 33, wherein the processing system is • The conversion module configured to convert the vibration of the movable member, and / or • At least one interference converter module for receiving sound interference, A voice generation system configured to reduce sound interference based on at least one electrical signal received from. Application Example 35: A voice generation system according to claim 33 or 34, wherein the sound interference in the conversion module includes a sound output by the speaker module and / or a phonation produced when the sound output by the speaker module is adjusted by the user's mouth movements. Application Example 36: It is a voice generation system, A housing having a first opening and a second opening, wherein the housing defines an air passage between the first opening and the second opening, A movable member is disposed within the housing and configured to vibrate in accordance with the flow of air in the air passage, A conversion module configured to convert the vibration of the movable member into an electrical signal, A speaker module configured to convert the aforementioned electrical signal into sound and output sound into the user's oral cavity, A processing system configured to reduce sound interference received by the aforementioned conversion module, A voice generation system equipped with the following features. Application Example 37: A voice generation system according to claim 36, wherein the processing system is • The conversion module configured to convert the vibration of the movable member, and / or • At least one interference converter module for receiving sound interference, A voice generation system configured to reduce sound interference based on at least one electrical signal received from. Application Example 38: A speech generation system according to claim 36 or 37, wherein the processing system is configured to reduce sound interference by performing feedforward suppression. Application Example 39: A voice generation system according to claim 38, wherein the feedforward suppression includes one or more of the following: gain reduction, frequency-selective noise reduction, notch filtering, phase modulation, and frequency shifting. Application Example 40: A speech generation system according to any one of claims 36 to 39, wherein the processing system is configured to reduce sound interference by performing adaptive feedback cancellation or residual feedback suppression. Application Example 41: A voice generation system according to claim 40, wherein adaptive feedback cancellation includes adaptive filtering. Application Example 42: A speech generation system according to claim 41, wherein adaptive filtering is performed by an FIR filter. Application Example 43: A speech generation system according to claim 41 or 42, wherein adaptive filtering is performed using a (real-time) normalized least squares mean (NLMS) algorithm. Application Example 44: A voice generation system according to any one of claims 36 to 43, further comprising at least one interference conversion module for receiving sound interference. Application Example 45: A voice generation system according to any one of claims 36 to 44, wherein the sound interference in the conversion module includes a sound output by the speaker module and / or a phonation produced when the sound output by the speaker module is adjusted by the user's mouth movements. Application Example 46: A sound generation system according to claim 45, further comprising a first interference converter module configured to convert the sound output by the speaker module into an electrical signal for input to the processing system. Application Example 47: A voice generation system according to claim 46, further comprising a second interference converter module configured to convert a phonation generated when the sound output by the speaker module is adjusted by the user's mouth movements into an electrical signal for input to the processing system. Application Example 48: A voice generation system according to any one of claims 36 to 47, wherein the processing system outputs at least one electrical signal to reduce sound interference. Application example 49: A voice generation system according to any one of claims 36 to 48, wherein the interference conversion module is • Microphone • Piezoelectric converter • Magnetic pickup converter ·Accelerometer • Voice sensor • Vibration sensor A voice generation system that includes one of the following: Application example 50: A voice generation system according to any one of the above claims, wherein the processing system is configured to process the electrical signal generated by the conversion module. Application Example 51: A voice generation system according to claim 50, wherein the processing system comprises hardware and software configured to improve the volume or quality of the voice encoded by the electrical signal. Application Example 52: A voice generation system according to claim 51, wherein the processing system comprises AI voice conversion software for improving the quality of the voice encoded by the electrical signal. Application Example 53: A voice generation system according to any one of the above claims, wherein the conversion module comprises a microphone. Application Example 54: A voice generation system according to any one of the above claims, wherein the conversion module comprises a piezoelectric converter. Application Example 55: A voice generation system according to any one of the above claims, wherein the conversion module comprises a magnetic pickup converter. Application Example 56: A voice generation system according to any one of the above claims, wherein the conversion module comprises a loudspeaker array. Application Example 57: A voice generation system according to any one of the above claims, wherein the conversion module is • Pressure sensor • Sound com • Vibration detection sensor ·Accelerometer A voice generation system that includes one of the following: Application Example 58: A sound generation system according to any one of the above claims, wherein the movable member comprises a membrane. Application example 59: A voice generation system according to any one of the above claims, wherein the movable member comprises a plurality of membranes having different degrees of complexity. Application example 60: It is a voice generation system, A housing having a first opening and a second opening, wherein the housing defines an air passage between the first opening and the second opening, A movable member is disposed within the housing and configured to vibrate in accordance with the flow of air in the air passage, A conversion module configured to convert the vibration of the movable member into an electrical signal, A speaker module configured to convert the aforementioned electrical signal into sound and output sound into the user's oral cavity, Equipped with, The output sound is generated by a voice generation system based on the air pressure inside the user's oral cavity. Application Example 61: A voice generation system according to claim 60, wherein the voice generation system is configured such that, during use, the air pressure at or near the second opening corresponds to the air pressure in the user's oral cavity. Application Example 62: A method for generating sound, A process of providing a movable member that interacts with an airflow corresponding to the airflow of the user's breathing, A step of detecting the vibration of the movable member in accordance with the airflow, A step of generating sound originating from the detected vibration using one or more electric speakers, A step of supplying airflow and sound to the user's oral cavity, A method that includes [a certain feature]. Application Example 63: A method according to claim 62, wherein the movable member is provided within an air passage, and the air passage is subjected to a first air pressure at a first end of the air passage and a second air pressure at a second end of the air passage. Application Example 64: A method according to claim 62 or 63, wherein the second air pressure corresponds to the air pressure in the user's oral cavity. Application Example 65: A method according to claim 62 or 63, wherein the first air pressure corresponds to the air pressure of the user's neck stoma. Application example 66: The method of claim 65, further, A step of detecting the air pressure and / or airflow in the user's neck stoma, A step of generating the first air pressure and the airflow of the neck stoma using an air pump, A method that includes [a certain feature]. Application Example 67: The method of claim 65, further, A step of detecting the air pressure in the user's oral cavity, A step of generating the second air pressure using an airflow control element, A method that includes [a certain feature]. Application Example 68: A method according to claim 67, wherein the airflow control element is an air pump. Application example 69: A method according to claim 67, wherein the airflow control element is an air valve. Application example 70: The method according to claim 67, wherein the airflow control element is ·Suction device Actuator A method that is one of the following. Application Example 71: A method according to any one of claims 67 to 69, further comprising the step of guiding the airflow of the user's breath in an air passage, wherein the airflow is the guided airflow of breath. Application example 72: A method according to any one of claims 62 to 71, further, A step of converting the vibration of the movable member into an electrical signal using a converter, The process of processing the aforementioned electrical signal, The process of supplying the processed electrical signal to the one or more electrical speakers in order to generate the sound, A method that includes [a certain feature]. Application example 73: A method according to claim 72, wherein the step of processing the electrical signal includes the step of improving the volume or quality of the sound encoded by the electrical signal. Application Example 74: A method according to any one of claims 62 to 73, further comprising the step of reducing sound interference generated by the one or more electric speakers.

Claims

1. It is a voice generation system, A housing having a first opening and a second opening, wherein the housing defines an air passage between the first opening and the second opening, A movable member is disposed within the housing and configured to vibrate in accordance with the flow of air in the air passage, A conversion module configured to convert the vibration of the movable member into an electrical signal, A speaker module configured to convert the aforementioned electrical signal into sound and output sound into the user's oral cavity, Equipped with, The voice generation system is configured such that, during use, the air pressure at or near the second opening corresponds to the air pressure inside the user's mouth.

2. A voice generation system according to claim 1, The second opening is configured to communicate with the user's oral cavity such that, when in use, the air pressure at or near the second opening corresponds to the air pressure inside the user's oral cavity, and / or The air pressure at or near the first opening corresponds to the air pressure in the user's neck stoma. A voice generation system wherein the first opening is optionally configured to communicate with the user's neck stoma such that the air pressure at the first opening corresponds to the air pressure at the user's neck stoma.

3. A voice generation system according to any one of claims 1 to 2, wherein the housing is configured to be connected to an airflow source, the airflow source is configured to generate an airflow and output the airflow into the user's oral cavity, and optionally, The airflow source is the user's neck stoma, and the airflow is the airflow of breathing output from the user through the neck stoma, or The aforementioned air source is an air pump, which is part of the sound generation system.

4. A voice generation system according to any one of claims 2 to 3, wherein the housing is configured to be attached to or around the user's neck such that the first opening communicates with the neck stoma.

5. The voice generation system according to claim 1, further, A pressure sensing module, which is attached to the user's neck and configured to detect the air pressure in the user's neck stoma, An air pump is disposed within the housing and configured to generate an airflow that moves along the air passage from the first opening to the second opening, A controller configured to control the air pump based on the detected air pressure in the neck stoma, Equipped with, The air pump is configured to generate air pressure in the air passage corresponding to the detected air pressure in the neck stoma, The air pump is a sound generation system that works in conjunction with the vibration of the movable member to drive the vibration of the movable member.

6. A sound generation system according to claim 5, wherein the speaker module is disposed within the housing and the speaker module is configured to output sound through the second opening.

7. The voice generation system according to claim 1, further, A first pressure detection module is attached to the user's neck and configured to detect the air pressure in the user's neck stoma. A second pressure detection module configured to detect the air pressure in or near the user's oral cavity, An air pump is disposed within the housing and configured to generate an airflow that moves along the air passage from the first opening to the second opening, A controller configured to control the air pump based on the detected air pressure in the neck stoma, the oral cavity, or a combination of both, is provided. The air pump works in conjunction with the vibration of the movable member to drive the vibration of the movable member. The second opening is configured to communicate with the user's oral cavity, and this is a voice generation system.

8. The voice generation system according to claim 1, further, A first opening configured to communicate with the user's neck stoma, A regulator, which is disposed within the housing and configured to control the airflow in the air passage, An air pump configured to generate an airflow into the user's oral cavity, A pressure sensing module configured to detect the air pressure inside the user's oral cavity, A controller configured to control the regulator based on the detected air pressure in the oral cavity, A voice generation system equipped with the following features.

9. A voice generation system according to claim 8, wherein the regulator is a further air pump configured to generate an air pressure in the air passage corresponding to the detected air pressure in the oral cavity.

10. A sound generation system according to claim 8 or 9, wherein the air pump and the movable member are disposed within a second housing configured to be fixed to the user's auricle, and the second housing includes an opening configured to communicate with the user's oral cavity, Optionally, the voice generation system further comprises a tube having a first end connected to the opening of the second housing, the second end of the tube being configured to be inserted into the user's oral cavity, and the pressure sensing module being attached to the second end of the tube.

11. A voice generation system according to any one of claims 8 to 10, wherein the air pump, the speaker module, and the pressure sensing module are attached to a denture unit configured to be fixed in the user's oral cavity.

12. A voice generation system according to any one of claims 1 to 11, further comprising a processing system, The processing system is configured to reduce sound interference received by the conversion module, Optionally, the processing system may The conversion module is configured to convert the vibration of the movable member, and / or At least one interference converter module for receiving sound interference, A voice generation system configured to reduce sound interference based on at least one electrical signal received from.

13. A voice generation system according to claim 12, wherein the sound interference in the conversion module includes the sound output by the speaker module and / or the utterance produced when the sound output by the speaker module is adjusted by the user's mouth movements.

14. A voice generation system according to any one of claims 1 to 13, wherein the voice generation system is configured to be used hands-free.

15. A method for generating sound, A step of providing a movable member that interacts with an airflow corresponding to the airflow of a user's breathing, wherein the movable member provided in an airflow channel is subjected to a first air pressure at the first end of the airflow channel and a second air pressure at the second end of the airflow channel. A step of detecting the vibration of the movable member in accordance with the airflow, A step of generating sound originating from the detected vibration using one or more electric speakers, A step of supplying airflow and sound to the user's oral cavity, Equipped with, The second air pressure corresponds to the air pressure inside the user's oral cavity, Optionally, further, A step of converting the vibration of the movable member into an electrical signal using a converter, The process of processing the aforementioned electrical signal, The process of supplying the processed electrical signal to the one or more electrical speakers in order to generate the sound, A method that includes [a certain feature].

Images