129 results about "Voice change" patented technology

The invention discloses a human-machine interactive voice control method and device based on a user emotional state, and a vehicle, wherein the method comprises: monitoring the expression, the voice or the motion of a set user; determining the current emotional state of the set user based on the expression, the voice or the action of the set user; determining the voice control mode of the vehicle according to the current emotional state of the set user; and performing the vehicle human-machine interaction according to the determined voice control mode. The method, the device and the vehicle can calculate the current emotion of the user according to the driving behavior, the speech speed and tone, and the facial expression of the user. An intelligent system can play appropriate music or adjust the navigation voice change according to the current emotional state of the user in order to achieve human-machine interaction with the user, thereby adjusting the user's emotion to achieve safer driving.

The invention relates to a sound-groove identification method based on emotion compensation. The emotion compensation includes three portions of emotion detection, character compensation and emotion expansion, comprising of calculating voice emotion factors to be according to the emotion detecting technique, compensating the voice change caused by emotion change respectively from the two layers of character and mode and finally improving robustness of the sound-groove identification technique to the emotion change. The invention has the advantages that the invention breaks through the inconsideration of sound-groove emotion change of the existing sound-groove identification technique, deals with the voice change caused by emotion change from the two layers of character and mode and strengthens resisting power to the voice emotion drift. The character layer standardizes the voice feature within the modeling ability of the training model by means of emotion degradation, normalization and barrier to reach the purpose of inhibiting the influence of the user emotion on the identification property. The mode layer obtains large scale emotion voices by employing the reverse way of synthesizing emotion voice by emotion changing rule, thereby greatly expanding the modeling information of the sound-groove model and resoling the difficulty of obtaining emotion data.

The invention discloses a voice changing method and device. The method comprises the following steps: receiving a source speaker statement; extracting voice recognition acoustic features and voice synthesis acoustic features from the source speaker statement; utilizing the voice recognition acoustic features to obtain voice recognition hidden layer features; utilizing the voice synthesis acousticfeatures to obtain voice synthesis coding features; inputting the voice recognition hidden layer features and the voice synthesis coding features into a pre-constructed tone conversion model corresponding to a specific target speaker to obtain voice synthesis acoustic features of the specific target speaker; and generating an audio signal of the specific target speaker by using the voice synthesisacoustic feature of the specific target speaker. According to the scheme of the invention, the voice changing method and device can achieve the conversion from any source speaker voice to the targetspeaker voice, and is better in voice changing effect.

The invention relates to a voice-changing system, a voice-changing method, a man-machine interaction system and a man-machine interaction method. The voice-changing system of the embodiment of the invention includes an audio input module which is used for receiving first audio information, a fundamental-frequency voice-changing module which is used for performing speed-changing tone-changing processing and / or speed-changing and tone-constant processing on the received first audio information so as to obtain second audio information, and an audio output module which is used for outputting the second audio information. According to the voice-changing system of the embodiment of the invention, tone-changing processing is performed on inputted audios, and therefore, the processing capacity of the voice-changing system can be enhanced, the problem of the monotony of an existing voice-changing system can be solved. According to the man-machine interaction system and method of the invention, visual sense, auditory sense and tactile sense are organically combined together so as to form a new interaction mode, and therefore, interactivity can be further improved.

The embodiment of the invention discloses a witness protection method and device for collecting audio / video evidence in real time. The method comprises the steps of carrying out face detection on video images of original videos which are collected in real time and are witnessed by witnesses; carrying out mosaic processing on face areas in the video images according to locations and sizes of the face areas if faces are detected, and encoding the video images after the mosaic processing; or otherwise, encoding the video images of the original videos; carrying out voice change processing on original audios which are collected in real time and are witnessed by the witnesses; encoding the audios after the voice change processing; synchronously synthesizing the encoded video images and the encoded audios into audios and videos; and sending the audios and videos to a non-evidence collection site device for play, thereby responding to the received evidence collection request information of the non-evidence collection site device. According to the scheme of the method and the device, the evidence collection efficiency is improved, and moreover, the identity leakage of the witnesses due to the fact that the mosaic areas are manually set not in time when the witnesses move relative to lenses is avoided.

The invention discloses a multifunctional reading machine, which comprises a power supply, a microprocessor, and a memorizer, a decoder, a digital-analog converter and an audio circuit connected with the microprocessor respectively. The reading machine is also provided with an automatic page turning mechanism, a camera shooting scanning device, a timer and a plurality of speech processors connected with the scanning device, a voice-change regulator and a digital voicing device. The memorizer comprises a one-off programming memory card, and the microprocessor processes and then outputs corresponding information through the decoder, the digital-analog converter and the audio circuit according to the scanned information. The reading machine can convert written information of printed reading matters, electronic publications and the like into voices to serve users, has the characteristics of strong anti-piracy ability, multiple languages, digital volume, word speed and tone regulations, small volume, convenient carry, flexible operation and the like, and has wide application range.

The invention discloses a cross-language timbre conversion system and method based on zero-order learning. The system sequentially comprises a mixed phoneme recognition module, a timbre conversion module, a speaker coding module and a vocoder module. According to the system, a voice signal Mel spectrum serves as an input signal, bottleneck features of the voice signal Mel spectrum are extracted through the phoneme recognition module, the features are normalized and then transmitted to an acoustic model, the Mel spectrum synthesized by the acoustic model is controlled by controlling a speaker reference vector, and finally audio is synthesized through a vocoder. The system can convert the voice of a common speaker into the timbre of a specified speaker, is suitable for accent corpora which do not appear in a training database, can be suitable for voice change of dialects in multiple regions, and has a wide application prospect.

The invention provides a voice simulation method and device. The method comprises steps that audio data of a user is acquired; the audio data is analyzed, and the characteristic information of the audio data is extracted and stored; simulation audio data corresponding to the audio data is generated according to the stored characteristic information; the simulation audio data is played. The method is advantaged in that voice is analyzed, characteristic data is extracted further, user interaction or reading is carried out through utilizing phonemes and a tone identical to those of the user, the sound simulation effect is good, high similarity is realized, the voice tones are similar, man-machine interaction cordial feeling is improved, and problems that only common voice change can be realized, sound can not be changed, a similarity level is low, man-machine interaction adaptability and cordial feeling can not be improved existing in a voice simulation method in the prior art are avoided.

The invention is applicable to the technical field of automatic speaker verification, and provides a voice change detection method and system, a mobile terminal and a storage medium. The method comprises the following steps of acquiring sample voice data, and carrying out feature extraction on the sample voice data to obtain a cqt voice feature; carrying out optimization processing on the cqt voice feature to obtain a cqcc voice feature, and inputting the cqcc voice feature into a preset convolutional neural network for model training in order to obtain a voice detection model; and acquiring to-be-detected voice, inputting the to-be-detected voice into the voice detection model for voice analysis, and carrying out voice change judgement on the to-be-detected voice according to an analysisresult of the voice detection model. According to the voice change detection method and system, the mobile terminal and the storage medium, the manual feature selection is not needed, the model training is carried out by adopting a convolutional neural network based mode, the accuracy of subsequent voice change detection for the to-be-detected voice is improved, and the resolution of the voice detection model is improved through extraction and optimization based on the cqt feature.

The invention provides an intelligent door lock. The intelligent door lock comprises a Bluetooth module, a processor, a voice processing module, a memorizer and a burning controller which are sequentially connected. The processor is connected with the burning controller. The intelligent door lock further comprises a voice playing module which is connected with the voice processing module. The invention further provides a voice burning method, a server and a storage medium. Due to the voice changing process of the intelligent door lock, user experience is improved.

The invention discloses a live broadcast control method and device, a storage medium and an electronic device. The method comprises the following steps: acquiring audio data to be shared through a client of a live broadcast application; in the process of collecting the audio data, acquiring a voice changing instruction generated by executing operation on an operation interface displayed in a client of the live broadcast application wherein the voice changing instruction carries a target voice changing type; and sending the target voice changing type and the audio data to a server of the live broadcast application, so that the server of the live broadcast application performs voice changing processing on the audio data according to the target voice changing type to obtain target audio datashared to the playing client for playing. The technical problem of relatively high live broadcast control operation complexity in a related live broadcast process is solved.

The invention relates to the technical field of electric cookers in household appliances, in particular to a voiceprint recognition controlled energy-saving electric cooker which aims at solving the structural problem that hard bottoms, sticky bottoms, electricity waste and the like often occur when the electric cookers are used for cooking in the prior art. The technical scheme includes that a vibration sensor is mounted on an electric cooker and connected with a voiceprint recognition circuit which is connected with a control circuit which is connected with a heating disc or a heating coil, and the like. A conventional control mode taking temperature changes of an inner pot as a control basis is changed into a control mode taking voice changes namely voiceprint changes inside the inner pot as a control basis, so that the problem that control lags behind actual temperature changes of the inner pot due to designability in the prior art of electric cookers. Voiceprints can reflect large amount of information, so that the control mode is more accurate and closer to actuality while electricity is greatly saved.

The invention relates to the technical field of audio devices, and particularly relates to an audio processing method and a system thereof which can improve the pleasant hearing enjoyment of a client.The method comprises the following steps: received audio signals are subjected to segmented processing to obtain noise samples, and the noise samples are stored one by one; after the audio signals are subjected to attenuation processing, the audio signals after attenuation processing and the stored noise samples are subjected to operational amplification processing, and audio signals after noisereduction are obtained; and the audio signals after noise reduction are sequentially subjected to sound change and amplification processing and are transmitted to an outer loudspeaking device. When the technical scheme provided by the invention is adopted, the audio signal quality is improved in the case of listening to a radio and receiving television network broadcast, and the sound speed and the tone are changed according to the habit of the client.

The invention provides a voice change conversation method, a device and a terminal. The method comprises the following steps: several voice change modes are established, and a corresponding trigger mode is established for each voice change mode; and during the communication process, whether some trigger mode is input is detected, and a corresponding voice changed mode is invoked for communication according to the input trigger mode when the detection result is that some trigger mode is input. By establishing the correspondence between the voice change modes and the trigger modes, different voice changed modes are switched during the communication process through the trigger modes. Thus, the problem that an existing voice change conversation scheme only can adopt a fixed voice change scheme and repeated voice changes cannot be realized according to different situations can be effectively solved.

The invention discloses a voice change detection method, a terminal and a computer readable storage medium. The method comprises the following steps of when a detection request is received, obtaininginformation of an object to be detected; detecting whether the object to be detected conforms to a corresponding preset condition or not; if so, obtaining corresponding data of voice to be detected; detecting whether the data of the voice to be detected conforms to the preset voice change detection voice condition or not; if so, obtaining corresponding feature information of voiceprints to be detected and voice counterfeiting judging results through a preset voice change detection model; detecting whether a preset voiceprint feature database is in a latest updated state or not; if so, obtaining the preset voiceprint feature information corresponding to the feature information of the voiceprints to be detected; calculating the matching degree between the feature information of the voiceprint features to be detected and the preset voiceprint feature information; and determining whether the voice data to be detected is artificially counterfeited voice data or not. Therefore, the technicalproblem of low detection accuracy of the artificially counterfeited voice is solved, and the detection accuracy of the voice data to be detected is improved.

The invention provides a method for integration and application of a voice changing system and a network telephone; a system which is combined by the voice changing system, an encoder, a decoder, and the like, is integrated in a network telephone application product which is self-developed, wherein, input voice is modulated instantaneously by the voice changing system, a modulated signal is sent into the encoder for encoding, is transmitted to a remote end through a network and decoded into a playable PCM signal format by the remote decoder, and the voice is instantaneously output by an audio device, therefore, the modulated voice can be heard.

The invention discloses a voice change method and device based on a specific target person voice change ratio parameter. The method comprises the steps of: obtaining the same content voice sample of a user and a specific target person; obtaining poles of a system respectively according to a sound channel modeling model, and deriving a voice change ratio parameter between the voices of the user and the specific target person; inputting a voice to be changed of the user, and moving poles according to the modeling model and the voice change ratio parameter, and obtaining a new pronouncing system model; and finally, utilizing the voice change ratio parameter to correct a pitch period excited by the voice to be changed, and restoring and outputting a changed voice signal in a new sound channel system. The voice change device has advantages that the feasibility is high, the installation is simple, the device cost is low, and the voice change device is applicable to various voice change application scenes.

The invention relates to a voice replacement method. The method comprises the steps of determining a replaced person in an audio / video resource, wherein the audio / video resource is a resource comprising audio information and image information, or the resource only comprising the image information, or the resource only comprising the audio information; determining an appointed person; obtaining audio information of the appointed person; and playing each frame of the audio / video resource in sequence, wherein for any frame, a playing mode comprises the fact that if any frame comprises the audio information corresponding to the replaced person, the audio information corresponding to the replaced person is replaced by the audio information of the appointed person, and then the audio replaced frame is played; if the any frame does not comprise the audio information corresponding to the replaced person and comprises the image information corresponding to the replaced person, the image information corresponding to the replaced person in the any frame is played, and moreover, the audio information of the appointed person is played; and if the any frame does not comprise the audio information corresponding to the replaced person and also does not comprise the image information corresponding to the replaced person, the frame is directly played. Person voice change after the audio / video resource is produced is realized, participation and interactivity and are improved.

The invention relates to a speech interaction method and a system applied in an intelligent door bell. The system comprises a speech acquisition terminal of the intelligent door bell, which acquires speech. The voice acquisition terminal of the intelligent door bell performs preset processing on the voice to obtain the voice after the first processing; a voice playing end of the intelligent door bell, receiving the first processed voice sent by the voice collecting end, and obtaining a second processed voice according to the first processed voice. The voice playing end of the intelligent doorbell plays the voice after the second processing; Wherein at least one of the first processed speech and the second processed speech is a voice after sound change. The present application eliminates potential safety hazards by preventing information such as gender, age, etc. of indoor personnel from being disclosed to visitors even if there is no indoor personnel or only vulnerable persons such asthe elderly and children.

The invention provides a microphone connection live broadcast method and related equipment, and the method comprises the steps: obtaining an original audio inputted by an anchor in real time based ona terminal and a target tone selected by the anchor based on the terminal if any terminal triggers a sound changing live broadcast mode in a live broadcast process of microphone connection of a plurality of terminals; performing tone conversion on the original tone in the original audio based on the target tone to obtain a converted target audio; and mixing the target audio with acquired originalaudios input by other microphone-connected terminals to obtain a mixed stream audio, and sending the mixed stream audio to all microphone-connected terminals and audience terminals entering a microphone-connected live broadcast room. In the scheme, a server performs tone conversion on an original audio input by a terminal triggering a voice-changing live broadcast mode in real time to obtain a target audio. Therefore, audiences entering the live broadcast room can watch conveniently. Through adoption of the method, microphone connection live broadcast is performed, so that the live broadcast watching experience of the user can be improved, and the user stickiness to the live broadcast platform is enhanced.

The invention discloses a voice change communication realization method. The method includes: receiving a call request inputted by a user, wherein the call request includes an interactive number of a current listened internet radio; judging whether the interactive number of the current listened internet radio exists in the voice change protection list or not; if yes, sending a voice change communication request to an internet radio server for the internet radio server to convert source voice of the user into target voice according to the voice change communication request, playing the target voice through the called terminal corresponding to the interactive number of the current listened internet radio. The invention further discloses a voice change communication realization terminal. By the method and the terminal, voice change communication can be realized during dialogue interaction with the internet radio, and user privacy is protected.

The application provides a call control method, a call control device, a mobile terminal and a computer readable storage medium. The call control method comprises the following steps: displaying a first voice change control on a call interface of the mobile terminal; when a first voice change instruction input based on the first voice change control is received, determining a voice change mode that is currently adopted under the indication of the first voice change instruction; performing voice change processing for a voice signal of an opposite end in a current call based on the currently-adopted voice change mode; and outputting the voice signal of the opposite end subjected to voice change processing as the voice signal of the opposite end in the current call. According to the technical scheme provided by the application, the modes that can be operated by users in a call process can be enriched.

The invention discloses a real-time digital voice changing method which comprises the following steps: adjusting and analyzing a non-unvoiced sound part of original voice, extracting a signal from a specific person fundamental tone base to replace an original fundamental tone according to a comparison result, and further performing synthesis and superimposition, so as to obtain a voice changing signal. According to the method, the voice changing effect has the characteristics of being high in naturalness and intelligibility; the change d voice is not recovered easily, so that the method is high in confidentiality; meanwhile, the method has the characteristics of being low in delay and complexity.

The invention provides a communication method used for a social network. The communication method comprises steps of receiving a voice which is sent to a second registered user by a first registered user in the social network; performing voice change treatment of the voice to generate a changed voice; storing the changed voice by a server; and presenting a listen interface of the changed voice in a computer graphical interface controlled by the second registered user, and hiding identity information of the first registered user from the second registered user. Correspondingly, the invention also provides a communication system used for the social network. By enforcement of the invention, the usage experience of the user in the usage of the social network can be increased by change in voice.

The invention relates to a speech sound function detection device. It includes signal generator, switching device, and voice signal probe unit. The switching device is set many switch to connect the input and output end of the speech sound module. The voice signal is inputted into the speech sound module by the given input end according to the demand in actual detection, and outputted by the given output end. The signal generator includes oscillator, frequency divider, low pass filter and attenuator. After the base frequency signal generated by the oscillator is divided frequency by the frequency divider, fed in the low pass filter to filter the waves, then voice signal with different frequency is generated. The voice signal is declined with different degree by the attenuator, which means the voice change can be simulated. The voice signal probe unit includes peak valve probe unit and frequency probe unit which are used to detect the peak valve and the frequency of the voice signal. The advantage are low cost, simple operation.

The invention provides a disease diagnosis system based on user voice change and a household intelligent robot. The disease diagnosis system based on the user voice change comprises a voice receivingmodule used for obtaining voice information of a target user, a timbre extraction module used for extracting current timbre characteristics of the target user according to the voice information, and an intelligent doctor module used for comparing the current timbre characteristics with pre-stored standard timbre characteristics of the target user and outputting diagnosis suggestions. According tothe invention, the current timbre characteristics of the target user are extracted through the timbre extraction module, then the current timbre characteristics are compared with the pre-stored standard timbre characteristics through the intelligent doctor module, and the diagnosis suggestions are output. The user can carry out active intervention at the early stage of illness or when the user does not feel the disease, so that the disease is restrained or treated in time, and the pain is reduced.

The invention discloses a frequency domain voice blind separation method for multi-frequency-band switching call media node (CMN) nonlinear function, which belongs to the technical field of speech enhancement and is characterized in that frequency domain speech is divided into the two frequency bands of low frequency and middle frequency based on kurtosis distribution characters. Three types of multi-frequency-band schemes are applied for switching a nonlinear function of a plurality of CMN algorithms, at least one scheme is led to be most matched with Gaussian performance and symmetry of frequency domain speech. Compared with single nonlinear function CMN algorithms, the frequency domain voice blind separation method for multi-frequency-band switching CMN nonlinear function is capable ofbeing adapted to voice changes in terms of Gaussian performance and symmetry, and remarkably improves voice separation performance. When an ordinary amplitude correlation method is adopted for voice sequence regulation of all frequency points, the separation signal to noise ratio of two paths of voice can be increased by 11dB at most, and the frequency domain voice blind separation method is stable is performance, easy in software and hardware achievement, and capable of being widely used in key technologies of computer perception and decision-making, unmanned driving and the like so as to achieve the speech enhancement function, and further improves entire performance of voice signal processing tasks such as voice recognition and content understanding.

The invention discloses an emotion recognition method. The method comprises the steps of receiving original recognition data sent by a terminal; recognizing the original recognition data to obtain voice feature data and face feature data, the voice feature data including voice feature time information and the face feature data including face feature time information; matching the voice feature data with a voice standard emotion model in an emotion model library to obtain voice change data; matching the face feature data with a face standard emotion model in an emotion model library to obtain face emotion change data; and verifying the voice change data according to the human face emotion change data, the voice feature time information and the human face feature time information to obtain an emotion recognition result. The invention further discloses a system and a computer readable storage medium. According to the invention, the change of user emotion can be identified, and the accuracy of user emotion recognition is improved.

The embodiment of the invention provides a voice conversion model and a training method thereof, and a voice conversion method and system. The training method comprises steps that a classification network model is trained through employing first sample data, and the first sample data to comprise a first audio and a first phoneme label corresponding to the first audio, the classification network model comprises a convolutional neural network layer and a recurrent neural network layer; second sample data are inputted into the trained classification network model, a second phoneme label corresponding to a second audio is obtained, and the second sample data comprise the second audio; the second audio frequency and the corresponding second phoneme label are used for training a voice changing network model, and the voice changing network model comprises a generator, a time domain discriminator and a frequency domain discriminator.

The embodiment of the invention provides a voice change recognition method and electronic equipment. The method comprises the following steps of obtaining target voice data; preprocessing the target voice data to obtain a voice signal sequence, wherein the voice signal sequence includes at least two sound signals and feature values of each voice signal, and the feature values at least include fundamental tone and resonance peaks; inputting the voice signal sequence into a preset voice change recognition model; and outputting a recognition result, wherein the recognition result is used for indicating whether the target voice data is subjected to voice change processing or not. The embodiment of the invention has the advantages that the voice change recognition on the target voice data is realized; and bad consequences caused when the voice change function is applied to abnormal scenes are avoided.