97 results about "Vocal intensity" patented technology

The invention relates an abnormal voice detecting method based on time-domain and frequency-domain analysis. The method includes computing the background sound intensity of a monitored scene updated in real time at first, and detecting and extracting suddenly changed fragments of the sound intensity; then extracting uniform filter Mel frequency cepstrum coefficients of the suddenly changed fragments; and finally using the extracted Mel frequency cepstrum coefficients of sound of the abnormal fragments as observation sequences, inputting a trained modified hidden Markov process model, and analyzing whether the abnormal fragments are abnormal voice or not according to frequency characteristics of voice. Time sequence correlation is improved when the hidden Markov process model is added. The method is combined with time-domain extraction of suddenly changed energy frames and verification within a frequency-domain range, the abnormal voice can be effectively detected, instantaneity is good, noise resistance is high, and robustness is fine.

The invention discloses an audio processing system and method for mobile terminals. The system comprises a first processing module, a judgment module and a second processing module, wherein the first processing module is used for carrying out low-frequency cutting operation on a received audio signal; the judgment module is used for making judgment according to the sound intensity of the audio signal, wherein the audio signal is judged as an effective signal if the sound intensity of the audio signal is greater than a preset threshold, and the audio signal is judged as ineffective signal if the sound intensity of the audio signal is lower than the preset threshold; and the second processing module is used for enhancing the volume of the effective signal according to a preset proportion and decreasing the volume of the ineffective signal according to a preset proportion or completely cutting the ineffective signal. According to the invention, the tone quality of records can be effectively improved, and tones become clear and full.

Provided are an automatic interpretation system and method for generating a synthetic sound having characteristics similar to those of an original speaker's voice. The automatic interpretation system for generating a synthetic sound having characteristics similar to those of an original speaker's voice includes a speech recognition module configured to generate text data by performing speech recognition for an original speech signal of an original speaker and extract at least one piece of characteristic information among pitch information, vocal intensity information, speech speed information, and vocal tract characteristic information of the original speech, an automatic translation module configured to generate a synthesis-target translation by translating the text data, and a speech synthesis module configured to generate a synthetic sound of the synthesis-target translation.

A device to monitor vocal information such as vocal volume and frequency and provide such information in a plurality of forms, such as tactile, audible, temporal, and visual; and to induce altered vocal volume and/or frequency by altering how the user hears his or her voice, such as by shifting the frequency at which the user hears his or her own voice; for the treatment of vocal nodules, Parkinson's disease, or stuttering.

The invention provides a method and device for controlling emotion. The method comprises the steps that 1, pulse frequency is acquired; 2, when a value of the pulse frequency is larger than a preset frequency value, acoustic information is acquired; 3, the acoustic intensity and beat speed in the acoustic information are detected; 4, whether a user is in an emotion-uncontrolled state or not is judged according to the acoustic intensity and beat speed; 5, when the user is in the emotion-uncontrolled state, acoustic information acquisition is stopped, and a preset audio is played. In this way, the problems that the emotion-uncontrolled state is automatically detected and emotion control is automatically performed according to the state are solved.

The invention relates to an acoustic imaging method, which solves the problems that general noise detection needs large computing power and cannot be adjusted to adapt to systems with different computing power conditions. The method comprises the following steps of: acquiring original audio data through a microphone; performing data processing on the original audio data to obtain a sound source intensity distribution diagram; collecting image information through a camera; and fusing the sound source intensity distribution diagram with the image information to obtain an acoustic imaging picture. The beneficial effects of the invention are that: the method achieves the visualization and materialization of the sound intensity through the combination of the sound source intensity distributiondiagram and the photographed image; a correction matrix is adopted to calibrate and correct the calculation deviation of the sound source intensity distribution matrix, and unnecessary interference iseliminated, so that the result is more precise; high-power interpolation processing is carried out after FFT processing, so that the computing power pressure is reduced, and meanwhile, the resolutionratio is also ensured; the simplified and optimized process can also be realized by using a small and light carrier device, and can be suitable for more application scenes.

Disclosed is a graphical user interface operational on a portable device that allows users to test their minimum audio hearing level at various frequencies using a single button. The graphical user interface generates sound frequencies that are played to the user, one ear at a time. Starting from zero decibels, sound intensity levels are gradually increased to intensity levels perceptible by the user. The users indicate recognition of the sound frequency by interacting with the graphical user interface. The intensities of the sound pressure levels (SPLs) are registered as estimates of a user's hearing sensitivity at the tested frequency bands. The process is repeated at different frequency bands for each ear, and hearing levels for each ear established.

Provided are a detection method for a sound source and a detection device therefor, which are capable of accurately detecting the position of a sound source through use of measurement results of sound intensities. In the detection method, I_all=√(I_x²+I_y²+I_z²), which is a total value of the sound intensities I_x, I_y, and I_z in three axial directions (x-, y-, and z-axis directions) orthogonal to each other at a sound receiving point (P) in a sound field, is calculated, and then the position of the sound source is estimated after extracting the sound source by performing processing with averaged values within a predetermined peak width instead of the magnitude of a peak in a sound intensity waveform of the total value, or the position of the sound source is estimated after extracting the sound source with attention being given to a spatial travel speed of a sound intensity or a sound particle velocity of the total value.

The invention discloses a method and a device for collecting voice information. The method includes: collecting a voice sample in a preset sampling period, and judging whether voice intensity corresponding to the voice sample reaches preset voice intensity or not; if yes, continuously collecting voice, and storing the same to analyze voice features in a sleeping state according to the stored voice. According to the method, the voice sample is collected, whether the voice sample meets collecting standards or not is judged, and if yes, continuous collection and storage are started, so that the voice features in the sleeping state can be analyzed according to the stored voice; an accurate collecting method is provided for users, convenience is brought to the users to know own states during sleeping, and the problem of poor user experience caused by the fact that a mobile terminal cannot maintain data collection and analysis in the sleeping process of several hours long in the prior art is solved.

The invention relates to the technical field of communication, in particular to a sound source positioning method and device, and the method comprises the steps: determining the time difference between the first receiving time when a sound source is collected through a first microphone and the second receiving time when the sound source is collected through a second microphone; determining reference position information of the sound source according to the time difference and a preset distance between a first microphone and a second microphone; determining an intensity difference between a first sound intensity of the sound source acquired by the first microphone and a second sound intensity of the sound source acquired by the second microphone; and determining target position informationof the sound source from the reference position information according to the intensity difference, the reference position information and a preset association relationship between the intensity difference and the position information, so the accuracy of sound source positioning can be improved.

The embodiment of the invention discloses a voice input method and device, electronic equipment and a storage medium. The method comprises the steps that when a voice input instruction is received, avoice signal collected by a microphone is received; obtaining environmental noise from the voice signal; when the sound intensity of the environmental noise is greater than a preset intensity threshold, obtaining a lip image; obtaining user input information corresponding to the lip image; and displaying the content corresponding to the user input information. According to the embodiment of the invention, noise interference is monitored in real time in a voice input process; the lip image of the user is acquired when the interference is too large, so that the input of the user is continuouslyacquired through lip language recognition, the user can still continuously perform input under the condition of noise interference, the risk of information leakage possibly caused by voice input of the user can also be reduced in time, and the information safety of the user is improved.

The embodiment of the invention provides electronic equipment, a signal processing method and device, a sound production device with two sound production units is arranged in the electronic equipment,a processor processes a first call audio signal to obtain two paths of sound signals, namely a second call audio signal and an offset audio signal. The first sounding unit is used for sending a callsound signal according to a second call audio signal, and the second sounding unit is used for sending a counteracting sound signal according to a counteracting audio signal, so that the counteractingsound signal and the call sound signal are mutually superposed outside the electronic equipment, and the sound intensity entering ears is not reduced as much as possible. Some signals, leaked out ofears, of the call sound signals can be counteracted through the counteracting sound signals, so that the sound leakage phenomenon of the electronic equipment is reduced, and the user experience is improved.

The invention discloses a method for controlling a hearing aid earphone, which comprises the following steps of: receiving an audio signal, and identifying the signal type of the audio signal; switching to a working mode corresponding to the signal type; and gaining according to a preset auxiliary hearing gain level corresponding to the working mode. According to the invention, when the earphone works in a certain working mode, the earphone can be controlled through the preset auxiliary hearing gain level corresponding to the working mode, so that the sound intensity played by the earphone can better meet the user demand, and the auditory experience of the user is improved; and the invention also provides a hearing aid earphone, a control device and system thereof and a computer readable storage medium, which have the same beneficial effects.

Disclosed is a ring shaped wearable communication device (ring) for a user's finger. The ring includes: a sound sensor that detects ambient sound levels; a speaker that produces sound in human audible frequency range per at least one signal received, wherein intensity of the sound produced is based on the ambient sound level; an out port; and a hollow channel with an increasing cross-section that wraps around the finger, wherein: smaller cross-section end of the hollow channel is connected to the speaker and the other end of the hollow channel is connected to the out port to provide a horn-like structure; the out port and the hollow channel are configured to reduce the mismatch between the acoustic impedance of air immediately outside the out port and the acoustic impedance as seen by the speaker just outside the speaker; and the out port emits sound received from the speaker as directional sound in close vicinity of the user's hand.

The invention discloses a volume adjustment method of a smart television, a storage medium and the smart television. The method comprises the following steps: determining target sound intensity of thesmart television; when it is detected that the user moves, determining a target watching distance between the user and the smart television; and determining a target volume value of the smart television according to the target sound intensity and the target watching distance, and adjusting the volume of the smart television to the target volume. According to the invention, different watching distances of the user are obtained, and the volume of the smart television is adjusted according to the target sound intensity, so that the user has the same auditory effect at different watching distances.

The invention discloses a shooting method and device and electronic equipment, and belongs to the technical field of camera shooting. The method comprises the following steps: under the condition that a shooting preview interface is displayed, acquiring voice information of a first voice signal , wherein the voice information comprises first sound intensity and sound source information; and under the condition that the sound source information indicates that the sounding object of the first voice signal is the object displayed in the shooting preview interface, executing zooming processing on a preview image in the shooting preview interface according to a first zooming magnification, wherein the first zoom ratio is associated with the first sound intensity.

The invention is suitable for the technical field of noise source positioning, and provides a noise source positioning method, device and system based on sound and images, and the method comprises the steps: obtaining sound signals collected at different positions; judging whether a loud noise position exists or not according to the sound intensity of the sound signal; if the loud noise position exists, controlling a camera to carry out image acquisition towards the loud noise position; and performing artificial intelligence algorithm identification on the image, judging whether personnel aggregation exists in the image, and if yes, determining that the position of personnel aggregation is the position of the noise source. According to the method, the device and the system, a sound and image combination mode is adopted, after the loud noise position is recognized according to the sound signal collected by the noise sensor, image collection and recognition are performed on the loud noise position through the camera, whether the noise source exists or not is further determined, and therefore the accuracy of noise source determination and positioning is further guaranteed.

The embodiment of the invention discloses a vehicle-mounted voice interaction method, device and equipment and a medium, and relates to the field of data processing, in particular to the voice technology. According to the specific implementation scheme, the method comprises the following steps: determining a target voice signal from at least two acquired voice signals according to the occurrence time of the voice signal and the sound intensity of the voice signal; denoising the target voice signal according to the sound source direction and waveform data of the target voice signal; and sendingthe denoised target voice signal to the target vehicle. The embodiment of the invention provides the vehicle-mounted voice interaction method and device, equipment and the medium, and aims to improvethe call quality of vehicle-mounted voice interaction.

A method is provided to measure a modulus of complex sensitivity of hydrophone data channels using a quasi-diffuse sound field. In the method, a radiation pattern representing shallow water sources directs to a location as a direct acoustic path and sound reflects to the location from the water bottom as a reflected acoustic path. At least one hydrophone receives the sounds at the location as acoustic signals with an acoustic intensity being the sound intensities along the acoustic paths. The sound intensity at the hydrophones also relates to a zenith angle and a bottom intensity reflection coefficient. The modulus of the frequency dependent sensitivity of the hydrophone is computed from measurements of the voltage output and voltages of reference hydrophones with the sound intensity as factor and with uncertainty reduced by averaging hydrophone sensitivities.

The embodiment of the invention discloses an audio key point determination method and device, equipment and a storage medium. The audio key point determination method comprises the steps that the feature point position of a target audio is determined, and feature points comprise a drumhead and a rhythm point; determining sound intensity corresponding to the feature points; and based on a preset key point determination rule, determining a key point of the target audio in combination with the sound intensity. According to the technical scheme of the embodiment of the invention, the feature points of the target audio, which comprises the drumbeat and the rhythm point, are utilized; the key points of the target audio are determined by combining the sound intensity of each feature point, so that the defect of inaccurate determination of the key points due to the fact that only a single feature point is used for determining the key points is overcome, and the effect of more accurately determining the key points of the audio is achieved.

The present invention discloses a pickup device and a cavity wall member for the pickup device, comprising: a housing having a pickup hole on a surface thereof; a microphone, wherein the microphone isarranged in the shell; a tunnel sound cavity which is formed in the shell, the microphone is formed in the shell, the tunnel sound cavity is formed in the shell and isolated between the pickup hole and the microphone, the pickup hole and the microphone are both communicated with the tunnel sound cavity, and sound waves entering the shell through the pickup hole are collected by the microphone through the tunnel sound cavity. According to the pickup device and the cavity wall component for the pickup device, the tunnel sound cavity is formed between the pickup hole and the microphone to increase the distance between the pickup hole and the microphone so as to achieve the effect of electrostatic isolation; and the loss of the sound intensity caused by the distance is compensated through theoscillation effect of the sound wave in the tunnel sound cavity.

The invention discloses an artificial intelligence speech recognition box for old people. The artificial intelligence speech recognition box comprises a box body; a signal conversion plate is fixedlyconnected to the interior of the box body; the surface of the signal conversion plate is fixedly connected with a hollow support spring; the top of the hollow supporting spring is fixedly connected with a signal receiving plate; the bottom of the signal receiving plate is fixedly connected with a vibrating rod; a metal plate is fixedly connected to the side, away from a pressure-sensitive switch,of an extrusion rod; a conductive rod is slidably connected to the side face of the metal plate; a fixing box is slidably connected to the side face of the extrusion rod; an air cylinder is fixedly connected to the side face of the hollow supporting spring, and a bearing air pipe is fixedly connected to the side face of the air cylinder. According to the artificial intelligence voice recognition box for old people, through cooperative use of the hollow supporting spring and a transmission gear, the effects that electric signals generated by sound waves are automatically improved when the speaking sound intensity is low, and the voice box can effectively receive instructions are achieved.

The invention discloses a noise reduction method and device, head-mounted display equipment and a readable storage medium, and the method comprises the following steps: obtaining a first audio signalreceived by a main microphone, and extracting first voice intensity and first noise intensity in the first audio signal; acquiring a second audio signal received by the auxiliary microphone, and extracting second voice intensity and second noise intensity in the second audio signal; comparing the first noise intensity with the second noise intensity to generate a proportionality coefficient; adjusting the sound intensity of the first audio signal or the second audio signal according to the proportionality coefficient to enable the noise intensity in the first audio signal and the noise intensity in the second audio signal to be the same; and processing the adjusted difference between the first audio signal and the second audio signal. According to the technical scheme, noise interference can be effectively eliminated, and voice distortion is avoided.

The invention provides a conference voice data processing method and system, and relates to the field of voice recognition. The method comprises the following steps: arranging a plurality of acquisition modules near different participants, and acquiring identity information and initial voiceprint features of the corresponding participants according to the different acquisition modules so as to acquire speech of the corresponding participants; identifying and judging whether the voice contents of the plurality of speech are the same, if so, analyzing the sound intensity of the plurality of voice contents, and selecting the voice content with the highest sound intensity; establishing a voice feature model of the plurality of participants according to the identity information and the initial voiceprint features, and inputting the selected speech into the voice feature model to obtain an identity matching result; and judging whether the identity information is matched with the identity matching result according to the acquisition modules, and if not, selecting the same voice content of the corresponding acquisition modules according to the identity matching result. The accuracy of voice acquisition of participants can be improved, and the conference recording effect is improved.