48results about How to "Voice enhancement" patented technology

The invention discloses a voice enhancing method, which comprises the following steps that voice signals with noises are subjected to noise reduction processing on the basis of a short-time spectrum amplitude estimation method; residue noise in the voice signals subjected to the noise reduction processing is subjected to shielding processing on the basis of the human ear shielding effect. The invention also discloses a voice enhancing device for realizing the voice enhancing method. Compared with the traditional voice enhancing method, the method and the device have the advantages that the processing step based on the human ear shielding effect is added on the basis of the short-time spectrum amplitude estimation method, so the residue noise can be shielded by the human ear, the problem of noise residue in the traditional voice enhancing method is effectively solved, and the effect of enhancing the voice is reached.

The invention discloses a headset communication method under a strong-noise environment and a headset. The method comprises: using earplugs to reduce medium and high frequency noises entering an ear canal, using an external connection cavity in parallel connection with the ear canal to divert medium and low frequency noises; using an internal microphone to pick up the sound in the ear canal and an environmental noise signal entering the ear canal, using an external microphone to pick up the environmental noise signal, and taking the external microphone signal as reference signals to eliminate the noise element in the internal microphone signal and remain the voice element to obtain transmitting terminal signals of the headset; using sound dynamic compression technology to cut down and compensate the signals picked up by the external microphone in terms of sound pressure level such that the sound pressure range is compressed to a range acceptable by human ears and the signals picked up by the external microphone and the receiving terminal signal received by the headset are broadcast together through a receiver of the headset. By means of the technical scheme of the present invention, the functions of protecting hearing, enhancing voice and monitoring a three-dimensional environment can be achieved comprehensively under strong-noise environments.

A method and system for enhancing a speech signal is provided herein. The method may include the following steps: obtaining an original video, wherein the original video includes a sequence of original input images showing a face of at least one human speaker, and an original soundtrack synchronized with said sequence of images; and processing, using a computer processor, the original video, to yield an enhanced speech signal of said at least one human speaker, by detecting sounds that are acoustically unrelated to the speech of the at least one human speaker, based on visual data derived from the sequence of original input images.

A signal processing apparatus processing a video signal and an audio signal in synchronization with the video signal includes generating means for generating information indicating a probability of a certain subject appearing in the image on the basis of the video signal that is input; determining means for determining whether the certain subject appears in the image on the basis of the information generated by the generating means; and directional characteristic varying means for, if the determining means determines that the certain subject appears in the image, varying a directional characteristic of the audio signal so as to increase the level of the audio signal collected from the direction of the subject and / or to decrease the levels of the audio signals collected from directions other than the direction of the subject.

The invention provides a speech enhancement method and system, computer equipment and a storage medium, and relates to the technical field of the human-machine speech interaction. The method comprisesthe following steps: collecting multi-channel acoustic signals through an acoustic vector sensor, preprocessing the multi-channel acoustic signals and acquiring a time-frequency spectrum, filtering the time-frequency spectrum and outputting a signal atlas; performing masking processing on the signal atlas through a nonlinear mask, and outputting an enhanced single-channel speech spectrogram; inputting the single-channel spectrogram into a deep neural network mask estimation model and outputting a mask spectrogram; performing time-frequency masking enhancement on the signal atlas through the mask spectrogram to acquire enhanced amplitude speech spectrogram; reconstructing through the enhanced amplitude speech spectrogram so as to output an enhanced target speech signal. The technical problem that the multi-channel speech enhancement is high in hardware cost, large in collection system volume, and high in operation complexity is solved, and the excellent speech enhancement effect can beacquired under difference interference noise types, strengths and room reverberation conditions.

A signal processing apparatus processing a video signal and an audio signal in synchronization with the video signal includes generating means for generating information indicating a probability of a certain subject appearing in the image on the basis of the video signal that is input; determining means for determining whether the certain subject appears in the image on the basis of the information generated by the generating means; and directional characteristic varying means for, if the determining means determines that the certain subject appears in the image, varying a directional characteristic of the audio signal so as to increase the level of the audio signal collected from the direction of the subject and / or to decrease the levels of the audio signals collected from directions other than the direction of the subject.

Methods and systems of identifying speech sound features within a speech sound are provided. The sound features may be identified using a multi-dimensional analysis that analyzes the time, frequency, and intensity at which a feature occurs within a speech sound, and the contribution of the feature to the sound. Information about sound features may be used to enhance spoken speech sounds to improve recognizability of the speech sounds by a listener.

The invention discloses a remote control method and a device of alarm. The method comprises: receiving alarm information reported by a base station and alarming on the spot according to the alarm information; and alarming remotely to a presupposed terminal under the condition that on the spot alarm is not received. Through the remote control method and the device of alarm, reliability and timeliness of the alarm are improved.

The present invention relates to a codec apparatus and method for coding / decoding speech and audio signals in a communication system. In accordance with the present invention, a speech and audio signal in a time domain is transformed into a speech and audio signal in a frequency domain and calculating frequency coefficients of the speech and audio signal, the frequency coefficients are split by a plurality of sub-bands and the sub-band coefficients of the respective sub-bands are calculated from the frequency coefficients, and the sub-band coefficients are quantized depending on a characteristic of the plurality of sub-bands and sub-band quantization indices are calculated by quantizing the sub-band coefficients.

The invention discloses a remote sound collection device and a remote sound collection method. The remote sound collection device comprises a sound pickup unit module, single-channel noise reducing processing modules, a microphone array processing module and a directive processing module; the sound pickup unit module comprises a reflective surface and a plurality of microphone components which arearranged in the center of the reflective surface; the output end of each sensor component is connected to the input end of the single-channel noise reducing processing module corresponding to the microphone component; the output end of each single-channel noise reducing processing module is connected to the input end of the microphone array processing module; and the output end of the microphonearray processing module is connected to the output end of the directive processing module.

A speech enhancement method operative for devices having limited available memory is described. The method is appropriate for very noisy environments and is capable of estimating the relative strengths of speech and noise components during both the presence as well as the absence of speech.

Systems and methods for detecting features in spoken speech and processing speech sounds based on the features are provided. One or more features may be identified in a speech sound. The speech sound may be modified to enhance or reduce the degree to which the feature affects the sound ultimately heard by a listener. Systems and methods according to embodiments of the invention may allow for automatic speech recognition devices that enhance detection and recognition of spoken sounds, such as by a user of a hearing aid or other device.

The invention relates to a mobile phone and a downlink voice processing method, comprising a baseband chip and a voice filtration module. The baseband chip sends the instructions of initialization and function setting to the voice filtration module; the voice filtration module receives the voice signals input by a transmitter and sends the voice signals to the baseband chip after processing the voice signals; the baseband chip judges whether to execute echo cancellation to the voice signals, and if yes, the baseband chip converts the voice signals to pulse code modulation (PCM )data and sends the data to the voice filtration module; and the data is processed with echo cancellation by the voice filtration module and is converted to voice signals which are then sent back to the baseband chip. Through treatment to the downlink voice signals, even in the case that the counterpart who is in conversation with the user of such functional mobile phone is in a background in which the noise increases suddenly, the voice of the speaker can be automatically increased and the noise can be reduced; the echoes can be eliminated or reserved; and specific noises can be eliminated, thereby realizing that voices can be clearly heard in a noisy environment or under special weather conditions.

A method for improving voice or video transmission quality in the cloud computing mode includes performing, by a cloud client, media negotiation with a communication peer end according to obtained media negotiation information of a corresponding local client; and establishing, according to a result of the media negotiation, a media channel between the local client and the communication peer end to perform voice or video transmission. An embodiment of the present invention further provides a corresponding cloud client and a corresponding local client. According to the method disclosed in the embodiments of the present invention, two client ends work collaboratively and a media channel is established on the local client, thereby ensuring voice or video transmission quality in the cloud computing mode.

The invention is applicable to the technical field of voice communication, and provides a voice processing method and device based on a generative adversarial network. The method comprises the following steps: acquiring voice training samples, wherein the voice training samples include N groups of complete voice samples, packet loss voice samples corresponding to the complete voice samples, K groups of broadband voice samples and narrowband voice samples corresponding to the broadband voice samples; putting the voice training samples into the generative adversarial network to carry out packetloss compensation model training based on the packet loss voice samples and the complete voice samples, and band spreading model training based on the broadband voice samples and the narrowband voicesamples, thereby obtaining a voice processing system composed of a packet loss compensation model and a band spreading model; and processing an original voice to be processed through the voice processing system to obtain an enhanced voice after packet loss compensation or band spreading. According to the voice processing method and device, the packet loss compensation processing efficiency based on a packet loss voice in voice processing, and the band spreading processing performance based on a narrowband voice can be improved.

A method and system for enhancing a speech signal is provided herein. The method may include the following steps: obtaining an original video, wherein the original video includes a sequence of original input images showing a face of at least one human speaker, and an original soundtrack synchronized with said sequence of images; and processing, using a computer processor, the original video, to yield an enhanced speech signal of said at least one human speaker, by detecting sounds that are acoustically unrelated to the speech of the at least one human speaker, based on visual data derived from the sequence of original input images.

The invention discloses a display teaching device used for marketing education. The display teaching device comprises a projection display device and a movable projection screen device. The projection display device comprises a main console and a supporting rod. A mobile table and a horizontal table are fixedly installed on the upper end of the supporting rod. A horizontal control rod is fixedly installed on the lower end of the horizontal table. A motor capable of controlling the vertical mobile table to elevate and controlling the horizontal table to move and a camera are fixedly installed on the lower end of the horizontal control rod. The right side of the main console is fixedly provided with a projector. The projector projects information to the movable projection screen device. The movable projection screen device comprises thin supporting plate type sound boxes which are arranged at the two sides and parallel, a movable projection screen and two layers of storage cabinets. According to the display teaching device used for marketing education, an object is photographed and projected by the projection display device and the movable projection screen device so that clear observation of the object can be realized and the teaching quality can be enhanced.

The invention discloses a headset communication method under a strong-noise environment and a headset. The method comprises: using earplugs to reduce medium and high frequency noises entering an ear canal, using an external connection cavity in parallel connection with the ear canal to divert medium and low frequency noises; using an internal microphone to pick up the sound in the ear canal and an environmental noise signal entering the ear canal, using an external microphone to pick up the environmental noise signal, and taking the external microphone signal as reference signals to eliminate the noise element in the internal microphone signal and remain the voice element to obtain transmitting terminal signals of the headset; using sound dynamic compression technology to cut down and compensate the signals picked up by the external microphone in terms of sound pressure level such that the sound pressure range is compressed to a range acceptable by human ears and the signals picked up by the external microphone and the receiving terminal signal received by the headset are broadcast together through a receiver of the headset. By means of the technical scheme of the present invention, the functions of protecting hearing, enhancing voice and monitoring a three-dimensional environment can be achieved comprehensively under strong-noise environments.

The present invention relates to a method for encoding a voice signal, a method for decoding a voice signal, and an apparatus using the same. The method for encoding the voice signal according to the present invention, includes the steps of: determining an eco-zone in a present frame; allocating bits for the present frame on the basis of the location of the eco-zone; and encoding the present frame using the allocated bits, wherein the step of allocating the bits allocates more bits in the section in which the eco-zone is located than in the section in which the eco-zone is not located.

A method and system for enhancing a speech signal is provided herein. The method may include the following steps: obtaining an original video, wherein the original video includes a sequence of original input images showing a face of at least one human speaker, and an original soundtrack synchronized with said sequence of images; and processing, using a computer processor, the original video, to yield an enhanced speech signal of said at least one human speaker, by detecting sounds that are acoustically unrelated to the speech of the at least one human speaker, based on visual data derived from the sequence of original input images.

The invention discloses a speaker speech enhancement method and device, and the method comprises the steps: extracting the features of a speaker from a registered audio, and carrying out the first processing of the features of the speaker, and obtaining the processed features of the speaker; performing second processing on the to-be-enhanced noisy voice to obtain a processed noisy voice; and splicing the processed speaker features and the processed noisy speech, and inputting the spliced speaker features and noisy speech into a speaker speech enhancement model for speaker speech enhancement. According to the method, the processed speaker features and the processed noisy speech are spliced and then input into the speaker speech enhancement model for speaker speech enhancement, so that the low-latitude speaker feature information can be fully used, and the speech of the speaker is further enhanced.

The invention relates to an intelligent cloud boxing-data collecting terminal. The terminal includes a shell body installed on the surface of a sandbag through ropes and an electric control module arranged on the shell body, wherein the electric control module is composed of a main control CPU, an acceleration sensor used for measuring direction of force conducted by a user and / or internal time of the force conducted by the user, and at least two left / right two-tone LED lights electrically connected to the main control CPU separately; the main control CPU is used for receiving a signal which is transmitted from the acceleration sensor and for measuring the direction of the force conducted by the user, and controlling the left / right two-tone LED lights in a corresponding direction to light up according to the signal. The main control CPU receives a signal which is transmitted from the acceleration sensor and for measuring intensity of the force conducted by the use; according to the strength of the signal, the main control CPU controls the luminance of the left / right two-tone LED lights in the corresponding direction. The intelligent cloud boxing-data collection terminal has the advantages of being reasonable in design, compact in structure and convenient to use.

The invention discloses a speaker recognition method based on sound-induced electroencephalogram signals. According to the method, by collecting electroencephalogram data, fusion features of time-frequency features and time-domain statistical features of an auditory stimulation part are extracted; fusion features obtained by electroencephalogram signals of an alpha frequency band baseline correction part are used as a background template; and the background template is subtracted from the auditory stimulation part fusion feature to obtain a clean task state data fusion feature, and finally different speakers are distinguished by using the network model provided by the invention. The invention provides a feasible speaker recognition method based on the sound-induced electroencephalogram signals, different speakers are distinguished by using the trained classifier, and the accuracy rate reaches 90%.

The invention discloses a microphone array voice enhancement system and method based on a multi-task network. The system is composed of a voice preprocessing module, a multi-task network module, a multi-task loss statistics module, a network weight calculation module and a voice reconstruction module. Wherein the voice preprocessing module acquires array voice, reference echo voice and target voice of each task as input voice and preprocesses the input voice; the multi-task network module completes reverberation removal, echo cancellation and noise reduction tasks of each sound channel of the array voice, fuses the multi-sound-channel voice and outputs the multi-sound-channel voice as enhanced voice; the multi-task loss statistics module is used for calculating the loss value of each task in the multi-task network module and counting the total loss of the network; the network weight calculation module calculates a gradient according to the total loss of the network, carries out back propagation on the gradient, and calculates the weight of the updated network; and the voice reconstruction module completes mapping from the frequency domain features to the time domain voice to obtain enhanced clean voice.

The invention discloses an audio optimization method, a related device, electronic equipment and a storage medium. The audio optimization method comprises the steps: extracting a first audio representation of a collected audio, and extracting a second audio representation of a reference audio; based on the first audio representation and the second audio representation, respectively extracting a first echo representation, a first voice representation and a first noise representation; performing interaction processing on the first voice representation, the first echo representation and the first noise representation to obtain a second voice representation, a second echo representation and a second noise representation, wherein the interactive processing comprises echo suppression, noise suppression and speech enhancement; and acquiring an optimized target audio based on at least one of the second speech representation, the second echo representation and the second noise representation. According to the method, the audio optimization effect can be improved.

Techniques for performing messaging between mobile stations (MSs) and UMA network controllers (UNCs) in an unlicensed mobile access network (UMAN). URR (UMA radio resource) messages are exchanged between an MS and one or more UNCs to perform various operations associated with UMAN. The MS may access the UMAN via a wireless access point (AP) that is communicatively coupled to the UNC via an IP network. The URR messages are sent between MSs and UNCs using an Up interface comprising a set of layered protocols over an underlying IP transport.

A controller selectively couples and de-couples at a rear surface of a portable information handling system and presents gaming content at both an integrated display and a peripheral display. The portable information handling system rests proximate the peripheral display when gaming content is presented at the peripheral display to present a communications interface of the gaming application. An end user manages gaming application communication between individuals and teams with gaze and voice inputs detected at the portable information handling system.

The invention discloses an echo cancellation method and device, equipment and a storage medium. The method comprises the steps of acquiring input signals of at least two microphones; processing the input signal based on a BF algorithm to obtain an output signal; and performing echo cancellation operation according to the input signals, the output signal and an echo cancellation algorithm to obtain a target output signal. According to the technical scheme of the invention, a method for further suppressing residual interference outside a beam by combining an adaptive echo cancellation technology is realized, so that voice in the beam is further enhanced, and a better effect of suppressing noise outside the beam is obtained.

This invention in its current embodiment provides for the wearer a mask that allows the wearer to breath, speak clearly with easy and hydrate at will while wearing the mask. The design of the mask is of 4 parts that assemble to make the perfect safety mask. As shown in FIG. 8, the mask is assembled by connecting the straw and noise connector along with the straw and voice box together and then assembling them together with part 1 as shown on FIG. 8 (the nose and mouth cover housing) which then makes it possible to bond them with the face cover shield labelled item 2 on FIG. 8. When all part are assembled together, the face mask becomes functional by allowing all who wear the mask to talk with clarity, hydrate at will without talking the mask off and breath with comfort.