452 results about "Silence" patented technology

The invention relates to the field of voice signal processing, and discloses a multi-level labeling method and device in a dialogue scene. Through the endpoint detection and automatic speech recognition method of the dialogue audio data, noise and mute in real speech can be removed, and effective speech segments can be obtained. And carrying out clustering analysis based on the effective voice segments, obtaining target voice segments and text data of the corresponding speakers, and carrying out emotion feature extraction through multi-modal emotion analysis. And finally, realizing automatic annotation of the inquiry strategy or intention understanding of the corresponding speaker by utilizing a large language model. According to the method, the recorded conversation voice can be subjected to multi-level labeling, an efficient and accurate information management and analysis tool is provided for a conversation scene, and the arrangement efficiency of conversation records can be greatly improved. The annotation information can be used for training and improving an artificial intelligence system, so that the artificial intelligence system plays a greater role in future psychological consultation service, and the overall quality and efficiency of psychological health service are improved.

The invention discloses a voiceprint recognition, identity confirmation and dialogue implementation method applied to sentiment analysis, which can well realize voiceprint recognition and identity confirmation, and can realize mute detection, recording control and recording end judgment and release. Comprising voiceprint recognition and identity confirmation: firstly, determining whether the source of recorded audio is the sound of a specific person by adopting a voiceprint recognition technology; the voiceprint recognition generally performs identity verification by extracting audio features and comparing the extracted audio features with a pre-stored template. The mathematical formula is as follows: assuming that the feature of the input audio is X = {x1, x2,..., xn} and the voiceprint template is T = {t1, t2,..., tm}, the recognition process can be represented as # imgabs0 #, and the # imgabs1 # represents the recognized user tag xj-ti2 represents the Euclidean distance between feature vectors. According to the method, sound cloning and emotion adjustment are realized through deep learning models (GANs and VAEs). In the whole process, through training and optimization of the deep neural network, high-quality emotional voice is generated, and more natural and personalized voice interaction experience is provided.

A metaverse application generates a first abuse score for the audio stream associated with a first user based on rules that analyze user signals, wherein the user signals include one or more selected from the group of one or more second users muting the first user, a history of past moderation actions associated with the first user, one or more third users blocking the first user, and combinations thereof. The metaverse application includes an audio-classifier model that obtains a second abuse score for an audio stream associated with the first user, wherein the audio stream is provided as input to the audio-classifier model. The metaverse application determines that the first user is committing abuse based on one or more selected from the group of the first abuse score, the second abuse score, and combinations thereof exceeding a threshold score value. Responsive to determining that the first user is committing abuse, the metaverse application performs a remedial action.

The utility model discloses a mute floor which comprises a surface treatment layer, a protection layer, a decoration layer, a first carrier, a first buffer layer and a second carrier. The surface treatment layer is arranged on one side of the protection layer, the decoration layer is arranged on the other side of the protection layer, the first carrier is arranged between the decoration layer and the first buffer layer, the first buffer layer is arranged between the first carrier and the second carrier, and the first buffer layer is of a porous structure and / or a net structure. A buffer groove is formed in at least one side face of the first buffer layer and extends to the two ends of the side face. By arranging the floor with the first buffer layer with the porous and / or honeycomb structure, noise is reflected and refracted for multiple times in the porous and / or honeycomb structure, so that sound is attenuated, a sound wave propagation path is prolonged, and the effect of more effectively reducing noise propagation is achieved.

The invention relates to the technical field of audio control, and discloses a distributed audio group mute and recovery method and system, a storage medium and a device, and the method comprises the steps: determining a master device in a distributed audio network, and enabling the master device to find and register a plurality of slave devices to establish an audio group; receiving a group mute or recovery instruction sent by a user through the master device, and performing validity verification on the instruction; the master device allocates a uniform timestamp to the verified instruction, and sends a synchronous control command to all slave devices in the group based on the timestamp; the slave equipment executes mute or recovery operation at the moment corresponding to the timestamp, synchronous response of the equipment in the group is realized, all the equipment is ensured to accurately and synchronously execute the mute / recovery operation through unified management of the master equipment and the adoption of global timestamp synchronization and adaptive delay prediction technologies, and an intelligent state monitoring and exception recovery mechanism is provided, so that the service life of the slave equipment is prolonged. Finally, efficient and reliable multi-room audio group control with excellent user experience is realized.

The embodiment of the invention discloses an audio processing method and device, and the method comprises the steps: carrying out the traversal of a segmented sentence list which is determined through the interception of an audio stream based on a mute detection result, and determining a current segmented sentence and an endpoint detection result of the current segmented sentence, determining a start frame of the current segmented sentence as a start frame of a latest to-be-identified sentence when the current segmented sentence has a sentence start point, determining an end frame of the current segmented sentence as an end frame of the latest to-be-identified sentence when the current segmented sentence has a sentence end point, and determining the latest to-be-identified sentence according to the start frame and the end frame of the latest to-be-identified sentence. According to the method and the device, the segmented sentences in the audio stream can be continuously updated, the to-be-recognized sentences can be determined according to the endpoint detection results of the segmented sentences, continuous long speech recognition is realized, and the complete content of the to-be-recognized sentences from the start frame to the end frame is recognized, so that the speech recognition efficiency is improved. The accuracy of long speech recognition can be improved.

The invention discloses an audio recognition and noise reduction processing method and system for a power grid regulation and control workbench environment, and the method comprises the steps: obtaining pure audio signals of all power grid regulation and control workers in different working scenes and different background noises in a dispatching desk, carrying out the cross-domain fusion of two features of a time domain and a frequency domain, and taking the two features as a training set to train a speaker model; voice signals are collected in real time, and pure noise and silence in the voice signals are removed; according to the noise level estimation and the noise power, carrying out weighting processing on the voice signal after the pure noise and the mute are removed; recognizing and separating human voice fragments by adopting a continuous down-sampling and resampling algorithm with multi-resolution characteristics; judging whether the human voice fragment contains a target speaker or not through a speaker model; and generating a plurality of voice segments based on the human voice segments, and extracting target speaking human voice through clustering and a speaker model. The method adapts to environmental changes of regulation and control dispatching desks at all levels, environmental noise can be remarkably reduced, and accurate recognition and extraction of human voices are achieved.

The invention provides a voice enhancement and high-precision recognition method and system in a complex environment, and relates to the technical field of voice processing, and the method comprises the steps: collecting a time domain signal in an off-road parking sentry box environment for preprocessing, detecting a mute segment signal in a standard time domain signal for noise power spectral density estimation, and obtaining a noise power spectral density value; a reverberation parameter is obtained by combining voice onset information and noise spatial correlation estimation, prediction is performed by using a deep neural network model, voice masking is applied to microphone array signals to perform enhancement processing, adaptive feature extraction is performed on time domain enhanced voice signals, and a voice signal is obtained. And performing high-precision recognition on the voice adaptive feature sequence based on an acoustic model and a language model, and outputting a target recognition text. The technical problems of poor voice signal quality and low recognition accuracy in a complex noise environment in the prior art are solved. The technical effects of improving the voice signal quality and the recognition accuracy and realizing clear, accurate and real-time voice interaction are achieved.

The invention relates to the technical field of wireless communication, in particular to an OTA upgrading method and system for Bluetooth sound equipment. The method comprises the following steps: acquiring to-be-upgraded firmware data and currently played audio data, and acquiring power amplifier current feedback data of a sound speaker by sending a double-frequency detection signal containing 21kHz and 22kHz; acquiring equipment transmission capability confirmation data according to a silent response mode of the sound equipment to the preset handshake signal; and calculating a nonlinear distortion coefficient of the loudspeaker according to the intensity of the 1kHz difference frequency component in the power amplifier current feedback data, and obtaining real-time channel evaluation data based on current waveform deviation analysis. According to the method, the nonlinear distortion degree of the loudspeaker is accurately evaluated through the 1kHz difference frequency component generated by the 21kHz and 22kHz dual-frequency detection signals, so that the reliability of data transmission is more accurately predicted.

The invention discloses a silent time window prediction method and device, a storage medium and an electronic device, and relates to the technical field of smart home, the method comprises the following steps: obtaining historical voice interaction data of a target object, the historical voice interaction data being used for indicating an interaction voice of the target object and a voice device; determining a first time difference between voice interaction time corresponding to historical voice interaction data and current time, and determining first sentence segmentation interval time corresponding to the historical voice interaction data; determining a target weight corresponding to the historical voice interaction data according to the first time difference; and predicting a target silence time window of interaction between the target object and the voice device according to the target weight and the first sentence segmentation interval time. Through the method, the problem that the overall voice interaction duration is increased due to the fact that a long silent time window is selected to adapt to different sentence segmentation habits of different user groups during speaking in the related technology can be solved.

The invention relates to the technical field of elderly care, solves the problems that in the prior art, due to the fact that noisy and intermittent elderly voice is lack of robust sentence bound recognition and interruption repair, misinformation and missing information of calling-for-help triggering are prone to occurring, and time delay is high, and provides an audio processing method and device for elderly calling-for-help, equipment and a medium. The method comprises the following steps: preprocessing a voice signal of the elderly to obtain a preprocessed voice signal; performing statement boundary recognition and labeling on the preprocessed voice signal to obtain a voice fragment set; carrying out interruption segment and silent interval detection on the voice segment set to obtain an initial voice segment sequence, and carrying out intelligent recombination and time sequence repair on the initial voice segment sequence to obtain a target voice segment sequence; and carrying out emergency call recognition on the target voice segment sequence to obtain an emergency call judgment result. According to the invention, the accuracy of distress call identification of the elderly is improved.

The invention relates to the technical field of voice interaction, in particular to a dialogue type AI low-delay response control method and system based on semantic analysis. The method comprises the following steps: firstly, generating a driving intensity correlation factor of a current time window according to the change of CAN bus data; further monitoring the output of the NLU, analyzing the acoustic intonation raising characteristics, and obtaining an interaction suppression coefficient in combination with the driving intensity correlation factor; further acquiring a time pressure coefficient based on the mute duration after the user stops sounding; further comparing the interaction suppression coefficient and the time pressure coefficient of the current time window to obtain a response judgment value; further analyzing the change trend of the interaction inhibition coefficient, and updating an environment improvement flag bit; and finally, according to the driving intensity correlation factor, the environment improvement flag bit and the response decision value of the current time window, carrying out multi-level response decision control, and carrying out selective reset of an interaction state, thereby solving the problems of wrong truncation and response delay of voice interaction under dynamic driving.

The invention relates to an audio recognition method and device based on marking and backtracking correction, and the method comprises the steps: segmenting a target audio, and enabling adjacent segments after segmentation to have a partial overlapping region; if a slice point exists in a non-mute segment of the target audio and the acoustic feature similarity of a preset number of frames before and after the slice point is smaller than a set similarity threshold value, the slice point is marked as a cut-off risk point, and the cut-off risk point is used for indicating that continuous semantics before and after the slice point has a cut-off risk; after a text corresponding to the audio of each slice is recognized through the speech recognition model, the language confidence degree of an overlapping area where the truncation risk point is located is determined, and the language confidence degree is used for indicating context language logic of the overlapping area; and if it is determined that the language confidence is lower than a set confidence threshold, correcting the recognition texts of the slices before and after the truncation risk point. According to the invention, the accuracy of audio recognition is improved.

An information processing system includes: a sensor unit that performs detection of contact of a pen-shaped input device with an operation target surface and presence of the pen-shaped input device within a detection limit distance on the operation target surface; a sound output unit that outputs a sound corresponding to input sound data; and a sound output control unit that inputs silent sound data to the sound output unit in response to the sensor unit detecting a first state in which the pen-shaped input device is detected within the detection limit distance, and inputs audible sound data to the sound output unit in response to the sensor unit detecting that, after the first state, a second state in which the pen-shaped input device is in contact with the operation target surface has been reached.

The present invention provides a soft decision audio decoding system for maintaining audio continuity in a digital wireless audio receiver, which infers the likelihood of errors in a received digital signal based on generated hard bits and soft bits. The soft audio decoder can utilize the soft bits to determine whether the digital signal should be decoded or muted. The soft bits can be generated based on a detection point and detected noise power or by using a soft output Viterbi algorithm. The value of the soft bits can indicate the confidence in the strength of the hard bit generation. The soft decision audio decoding system can infer errors and decode perceptually acceptable audio without error detection as in conventional systems and has low latency and improved granularity.

An apparatus comprising means configured to: obtain, from a further apparatus, at least one bitstream comprising at least one component or stream of an immersive audio bitstream; identify at least one of the at least one component or stream of the immersive audio bitstream to be muted; transmit, to a further apparatus, a mute request indicating the identified at least one component or stream to be muted; obtain, from the further apparatus, the at least one bitstream, wherein the at least one bitstream is modified with respect to the identified at least one component or stream of the at least one immersive audio bitstream to be muted, wherein the modification in the at least one bitstream, is at least one of: the at least one component or stream having reduced energy; the at least one component or stream not being received; the at least one component or stream comprises a silence descriptor indicator for indicating the at least one component or stream is to be muted; the at least one component or stream having a reduced bit rate; and at least one other component or stream having an increased bit rate.

Content aware audio processing includes receiving, by a digital signal processor, a frame of audio data. In response to detecting that the frame of audio data is a silent frame, the digital signal processor selects a light graph from a plurality of graphs including the light graph and a full graph. Comfort noise is generated that corresponds to the silent frame. The comfort noise frame is processed through the light graph in place of the silent frame. The light graph is dedicated for processing comfort noise frames.

An input audio signal associated with a communication session is received. A determination is made that a participant associated with the input audio signal is not audibly speaking within the input audio signal. In response to determining that the participant is not audibly speaking, an audio feed corresponding to the input audio signal is muted by rendering the audio feed not audible to at least one other participant device connected to the communication session.

The utility model relates to the technical field of silence bins, in particular to a silence bin assembling device which comprises a supporting bottom frame, a stand column is fixedly installed on the surface of the supporting bottom frame, a top frame is fixedly installed on the top of the stand column, a limiting mechanism is arranged on the surface of the stand column, and the limiting mechanism comprises a fixing support. The fixing support is fixedly installed on the surface of the stand column, a first sound insulation strip is fixedly connected to the surface of the fixing support, a side plate is connected to the surface of the first sound insulation strip in a contact mode, a spring is fixedly connected to the surface of the stand column, and a limiting rod is fixedly connected to the surface of the spring. According to the utility model, the side plate can be limited by the limiting rod only by pushing the side plate to the fixed bracket to extrude the first sound insulation strip, and the side plate can be mounted without manual support, so that the number of mounting personnel is reduced, the mounting cost is reduced, and the competitiveness of a product is further improved.

The invention provides lightweight sound-absorbing mute cotton and a preparation method thereof, the lightweight sound-absorbing mute cotton is of a bionic gradient structure and is divided into three layers, the bottom layer dissipates high-frequency sound waves, the middle layer optimizes intermediate-frequency resonance, and the upper layer enhances low-frequency sound wave scattering. The preparation method comprises the following steps: respectively preparing bottom-layer slurry, middle-layer slurry and upper-layer slurry; pre-cooling the bottom of a mold, and carrying out synchronous three-layer co-injection molding on the three kinds of slurry: carrying out one-way freezing to form a longitudinal gradient aperture; freezing and drying to obtain a three-layer gradient structure material; immersing in a graphene oxide quantum dot ethanol solution for vacuum treatment; heating, adding ascorbic acid, keeping for 2 hours, and applying 0.5% pre-strain to the material; photo-thermal curing is performed; and performing hot-pressing shaping to obtain the product. The gradient structure design is adopted, and the aperture, the fiber distribution and the filler type are regulated and controlled, so that acoustic impedance gradient change is realized, sound wave reflection is reduced, and the sound absorption efficiency is improved.

The utility model discloses a conjoined junction box with a sound insulation function. The conjoined junction box comprises a first junction box, a second junction box and a sound insulation body, the sound insulation body is arranged between the first junction box and the second junction box; the shells of the first junction box and the second junction box are also wrapped with a third noise reduction layer; and a fixing piece is arranged outside the one-piece wire box. According to the invention, by arranging the one-piece junction box in the household wall, the problems of repeated construction and noise generated in the construction process when the junction box is independently arranged in the household wall can be avoided, the damage to the structure of the household wall in the construction process is reduced, and in addition, the sound insulation body is arranged between the first junction box and the second junction box, so that the sound insulation effect is improved. In the process of installing the junction box on the household wall body, the mute effect in the installation process of the junction box at the same position between household splitting can be achieved, and the obvious noise reduction effect in the use process of the junction box in the household wall can be achieved.

This application discloses an intelligent method and apparatus for detecting active speech with sentence segmentation based on multi-state temporal modeling. The method includes: receiving audio signals from at least one channel; extracting acoustic feature sequences from the audio signals using a target speech recognition model corresponding to the number of channels; determining the probability distribution of each speech frame corresponding to the acoustic feature sequences belonging to different speech activity states, obtaining a state sequence corresponding to each channel, wherein the speech activity state includes at least one of the following: initial silence state, speech state, intra-turn pause silence state, and inter-turn sentence segmentation silence state; and determining the time of sentence segmentation in the audio signal based on the state sequence. This application solves the technical problem of erroneous sentence segmentation in speech activity detection based on a fixed silence threshold in related technologies.

A method includes obtaining initial transcription for input natural speech; performing segmentation of initial transcription into text portions, based on punctuation marks in initial transcription; determining segment-level timestamps for text portions based on the input natural speech; performing audio segmentation on input natural speech, by cutting input natural speech based on segment-level timestamps, to obtain audio chunks; generating transcription portions for each of the audio chunks; merging transcription portions to form re-transcription; determining word-level timestamps for re-transcription, by aligning input natural speech against re-transcription; calculating silence time periods, each corresponding to silence between each two adjacent words of input natural speech, based on word-level timestamps; performing a final segmentation on input natural speech and re-transcription, based on silence time periods, to generate final audio segments and corresponding final transcription portions. The final audio segments and corresponding final transcription portions may be included in training dataset for training a model.

The utility model discloses a low-noise intelligent oxygen cabin. The low-noise intelligent oxygen cabin comprises a cabin body, a cabin door, an inner isolation plate, an integrated control part, a monitoring sensor group and an operation part, the cabin body comprises a hollow shell made of sound insulation materials. The cabin door is made of sound insulation glass and can be installed at an access port of the cabin body in an openable mode. A sound attenuation and noise reduction isolation layer made of a sound absorption material is arranged in the inner isolation plate, and the inner isolation plate is detachably connected to the connecting step of the cabin body in a sealed mode; the monitoring sensor group monitors pressure, temperature and oxygen-enriched concentration data in real time and feeds back the pressure, temperature and oxygen-enriched concentration data to the integrated control part, and the integrated control part controls the oxygen generation module, the pressurization module and the temperature control module to work or start and stop, so that the silent oxygen chamber is at the set temperature, pressure and oxygen-enriched concentration. The utility model has the advantages of low noise, silence environment, immersion treatment of personnel in the cabin, constant-temperature, constant-pressure and constant-oxygen medical environment of the silence oxygen chamber, and good medical effect.

A method includes obtaining initial transcription for input natural speech; performing segmentation of initial transcription into text portions, based on punctuation marks in initial transcription; determining segment-level timestamps for text portions based on the input natural speech; performing audio segmentation on input natural speech, by cutting input natural speech based on segment-level timestamps, to obtain audio chunks; generating transcription portions for each of the audio chunks; merging transcription portions to form re-transcription; determining word-level timestamps for re-transcription, by aligning input natural speech against re-transcription; calculating silence time periods, each corresponding to silence between each two adjacent words of input natural speech, based on word-level timestamps; performing a final segmentation on input natural speech and re-transcription, based on silence time periods, to generate final audio segments and corresponding final transcription portions. The final audio segments and corresponding final transcription portions may be included in training dataset for training a model.

To predict whether there is no sound in a predetermined space based upon whether onomatopoeia indicative of silence that a human perceives in a quiet environment is generated.SOLUTION: A determination device comprises: an input part which inputs predetermined sound data generated by converting a predetermined sound into numeric data through frequency analysis to a determination model outputting whether a specific frequency of 1,800 Hz or more to 2,200 Hz or less indicative of induced otoacoustic emissions is included in addition to the frequency of sound data once the sound data is input; and a control part which displays, on a display unit, the generation of onomatopoeia indicative of silence that a human perceives in a quiet environment when the determination model receiving the predetermined sound data generates an output showing that the specified frequency is detected, but displays, on the display unit, no generation of onomatopoeia when the determination model generates an output showing no detection of the specific frequency.SELECTED DRAWING: Figure 1

The utility model discloses a double-broken-bridge double-inner casement window capable of improving low-frequency noise reduction capacity of doors and windows, which is provided with a frame, a screen sash, a glass sash and a window frame of the double-broken-bridge double-inner casement window formed by the frame. The frame material is of a tooth-shaped inner cavity structure, the design is different from a traditional window type, the tooth-shaped design mode is adopted, the low-frequency noise reduction capacity of the door and window is improved, and the material strength is improved. The double-broken-bridge double-inner casement window is in a heat-insulation anti-theft two-in-one double-inner casement window type, the width of a frame is 112 mm, and frame sashes are connected through movable glue injection corner connectors in a 45-degree corner combining mode. The glass screen sashes are opened towards the indoor direction and provided with stainless steel and diamond screens, so that the whole window is attractive in appearance, good in sealing performance and convenient to machine and manufacture.

The present application provides a method, apparatus, device and storage medium for voice endpoint detection, wherein the voice endpoint detection method can determine whether the audio frame contained in the audio data to be detected is a silence frame, a noise frame or a voice frame, that is, the present application can detect the relatively accurate attributes of the audio frame contained in the audio data, and perform detection of the voice front-end point and the voice back-end point on this basis, so as to obtain relatively accurate detection results. On the basis of realizing voice endpoint detection, the present application can obtain the recognition text of the voice segment, and can determine the semantic scene of the recognition text based on the semantics of the recognition text, and then can set a suitable post-silence timeout threshold based on the semantic scene of the recognition text, thereby triggering a post-silence timeout event based on the suitable post-silence timeout threshold to improve the user experience.

The invention relates to an AdobeAudition (AU) extension program-based automatic track matching system and method, and relates to the technical field of audio processing, a picture book analysis module identifies and analyzes a role line sequence and an interval parameter of a line book through a regular expression and a role tag; the audio matching and loading module is used for matching role audios according to a filename rule or a metadata label and loading the role audios to an initial track; the silence detection and marking module adds a mark at a non-silence initial position by adopting framing energy calculation; the line-audio verification module is used for realizing audio and line matching degree analysis through voice recognition to text and font pinyin double-path similarity calculation, and providing a visual calibration interface; and the automatic execution module supports a full-automatic sequential splicing mode or a semi-active sentence-by-sentence splicing mode. The method comprises the steps of line book analysis, audio matching and loading, mute detection marking, verification and calibration and dual-mode execution. The problems of low manual rail alignment efficiency and poor precision in the audio book making process are solved.

The invention discloses a multi-person voice processing system and separation method based on voiceprint recognition and clustering, and belongs to the technical field of multi-person voice separation processing, and the system comprises a voice activity detection unit, an audio cache unit, a voiceprint processing unit and a data processing unit. According to the method, voice activity detection, voiceprint technology, speaker separation technology and caching technology are fused, voice activity detection is performed on a section of voice, non-human voice such as silence, noise and music in the audio is removed, only human voice is reserved, the caching technology is used for caching the audio data, and when the cached data reach a preset threshold value, the voice data are cached. The method comprises the following steps: caching audio data, segmenting the cached audio data according to a certain frame shift and frame length, then extracting speaker features in sub-audio segments by using a voiceprint recognition technology, finally clustering the features, recognizing the audio segment of each person, and outputting corresponding time information.