Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

47results about How to "Implement speech recognition" patented technology

Voice recognition method facing specific crowd

The invention discloses a voice recognition method facing a specific crowd. The method comprises the following steps of: first, sampling a voice signal and converting the voice signal to a digital signal from an analogue signal; then, pre-weighting, windowing, en-framing and performing front-end processing of endpoint detection on the digital voice signal; later on, performing feature extraction on the voice signal by adopting discrete wavelet transform; and finally, performing voice recognition on the feature-extracted voice signal by adopting a discrete hidden Markov model after training a sample. In the processes of performing the front-end processing and the feature extraction on the voice signal, spectrum features and pronunciation characteristics of different target crowds are fully taken into consideration and the process of extracting voice information is optimized, so that a processing process and an information extracting process can be simplified; and therefore, recognition precision is ensured, simultaneously calculation amount and information storage capacity in the recognition process are greatly reduced, and the voice recognition on an embedded platform is realized.
Owner:HANGZHOU PINGPONG INTELLIGENT TECH CO LTD

Voice recognition method and device and electronic equipment

The invention relates to the technical field of voice recognition, provides a voice recognition method and device and electronic equipment, which aim to solve the problem of relatively low voice recognition accuracy. The method comprises the following steps of acquiring a to-be-recognized voice, performing feature extraction on the to-be-recognized voice to obtain voice feature information, determining a target character sequence corresponding to the voice feature information according to the target acoustic model and the target language model, wherein the target language model comprises a first language model and a second language model, the first language model is obtained by performing language model training through a command word training text of a first scene, and the second languagemodel is obtained by performing language model training through a first text training set. In the voice recognition process, two language models are adopted, and the first language model is obtainedby performing language model training through the command word training text of the first scene, so that the recognition capability of the first language model for related command words in the first scene can be enhanced, and the voice recognition accuracy can be improved.
Owner:SOUNDAI TECH CO LTD

Recording realization method and device in conversation process

InactiveCN105847520ASolve the problem of large recording filesImprove the efficiency of recordingSubstation equipmentSpeech recognitionTelecommunicationsComputer terminal
The invention relates to a recording realization method and device in conversation process. The method comprises the following steps: determining telephone number of the current communication is a preset telephone number; if the telephone number is the preset telephone number, determining whether the voice content of the current communication comprises a recording start keyword; and if the voice content comprises the recording start keyword, controlling a user terminal to start recording the current communication. The technical scheme can effectively control the user terminal to start the recording function automatically according to the communication content to record user-required information in the conversation process.
Owner:BEIJING XIAOMI MOBILE SOFTWARE CO LTD

Voice separating method based on auditory center system under multi-sound-source environment

The invention provides a voice separating method based on an auditory center system under the multi-sound-source environment, and relates to the field of digital signal processing. The voice separating method solves the boundedness that most voice recognition methods can only be used under the low-noise and single-sound-source environment. To carry out the voice recognition under the multi-sound-source noisy environment, voice separation needs to be achieved firstly. According to the voice separating method based on the auditory center system, the multi-spectra analysis is carried out on voice signals through a peripheral hearing model, a coinciding nerve cell comprises a general cynapse model and a general cell model to integrate the information of an ITD and the information of an ILD, the voice separation is achieved in a hypothalamus cell model, and the experiment shows that the method has good robustness.
Owner:CHONGQING UNIV OF POSTS & TELECOMM

Digital household system

This invention relates to digital house system composed of television display module and remoter, which is characterized by the following parts: the system comprises central processor, display control module, north bridge chip, memory, network module, south bridge chip, common bus, PCI bus, remote receive module, memory module, power control module, input and output module, sound input control module, sound integration module, house network gate, materials servo, remote control end, bedroom terminal, environment test module, household management module, house safety module, remote alarm module, cell phone module and area alarm center.
Owner:李大东

User-customization multi-mode general remote controller

The invention relates to a general remote controller and aims at providing a multifunctional general remote controller which supports multiple modes and can be defined according to user requirements. The general remote controller comprises a processor, an operation system, an infrared emitting tube, a radio frequency module, a ZigBee module, a radio frequency module group, a gyroscope, a gravity sensor, a displayer, a thin film transistor (TFT) capacitor screen, a lithium ion battery, a player and a sound pickup. The operation system is installed in the processor, and the infrared emitting tube, the radio frequency module, the ZigBee module, a radio frequency module group, the gyroscope, the gravity sensor, the displayer, the TFT capacitor screen, the lithium ion battery, the player and the sound pickup are respectively connected with the processor through serial input / output ports.
Owner:浙江达峰科技有限公司

Speech recognition model training method, speech recognition method and related devices

The embodiment of the invention provides a speech recognition model training method, a speech recognition method and related devices. The training method comprises the steps of: determining a trainingcurrent mixed language audio, obtaining the training initial acoustic features, utilizing a first language module to obtain a training first time sequence position acoustic feature, utilizing a second language module to obtain a training second time sequence position acoustic feature, performing fusion and text coding on the training first time sequence position acoustic feature and the trainingsecond time sequence position acoustic feature to obtain a training current fusion text feature, obtaining a first training current prediction text feature according to the training current fusion text feature and a previous reference text feature, obtaining first loss according to the first training current prediction text feature and a current reference text feature, then obtaining model loss, and adjusting the parameters of a speech recognition model according to the model loss until the trained speech recognition model is obtained. According to the voice recognition model training method,the voice recognition method and the related devices provided by the embodiment of the invention, the voice recognition accuracy can be improved.
Owner:BEIJING CENTURY TAL EDUCATION TECH CO LTD

Smart voice cell phone or smart voice tablet computer

The invention discloses a smart voice cell phone or smart voice tablet computer and relates to the technical field of electronic devices. The smart voice cell phone or smart voice tablet computer comprises a cloud server and an electronic device. A voice wake-up module, a voice recognition module, a voice command module, a semantic analysis module, a voice synthesis module and a program control module are arranged in the electronic device. The voice recognition module is connected with the cloud server through a wired communication module. The voice recognition module is connected with the voice command module through an information analysis and processing and signal conversion and transmission module. The voice command module is connected with the program control module through an information processing and signal transmission module. By providing the voice wake-up module, the voice recognition module, the voice command module, the semantic analysis module, the voice synthesis moduleand the program control module in the electronic device, the smart voice cell phone or smart voice tablet computer achieves voice interaction, directly wakes up programs through the voice wake-up module without manually opening the applications, realizes software application through a voice interaction function easy to operate.
Owner:ANHUI SEMXUM INFORMATION TECH CO LTD

Speech recognition of mobile telecommunication terminal

A method for identifying voice of mobile communication terminal includes establishing video ¿C audio model by utilizing triphone, registering telephone number, forming triphone data for registered telephone number name or firm name and storing it, inputting user voice, comparing inputted voice data with stored triphone data, dialing out telephone number of inputted voice if identification is successful.
Owner:LG ELECTRONIC (HUIZHOU) CO LTD

Voice recognizing method and device based on block chain, medium and electronic device

The embodiment of the invention provides a voice recognizing method and a voice recognizing device based on a block chain, a medium and an electronic device. The voice recognizing method based on theblock chain comprises the following steps: storing labelled historical voice data in the block chain; and if a new block of the current voice information is generated in the block chain, triggering anacoustic model and a linguistic model finishing through training to recognize the current voice information, and outputting the current text information corresponding to the current voice information, wherein the acoustic model and the linguistic model are obtained through training by adopting the historical voice data labelled in the block chain. According to the technical scheme of the embodiment of the invention, the voice data of the user is stored by adopting the block chain technology, and the recognition for the current novice is carried out based on the stored historical voice data.
Owner:TAIKANG LIFE INSURANCE CO LTD

Intelligent financial counseling robot convenient to use

InactiveCN108406782AHandling financial consulting services convenientlyImprove stabilityProgramme-controlled manipulatorComputer caseBevel gear
The invention discloses an intelligent financial counseling robot convenient to use and relates to the technical field of financial counseling equipment. The intelligent financial counseling robot convenient to use comprises a bottom plate. The top of the bottom plate is fixedly connected with a crate. A transmission case is connected between the two sides of the inner wall of the crate in a sliding mode. The bottom of the inner wall of the transmission case is fixedly connected with a motor through a connecting block, and the outer surface of an output shaft of the motor is fixedly connectedwith a first bevel gear. A bidirectional threaded rod is rotationally connected between the two sides of the inner wall of the transmission case through bearings, and the outer surface of the middle of the bidirectional threaded rod is fixedly connected with a second bevel gear adapting to the first bevel gear. According to the intelligent financial counseling robot convenient to use, people can use the robot quite conveniently, the height of a touch screen can be automatically adjusted according to the height of a client and the distance to the client, the purpose that the robot is suitable for different clients is well achieved, and accordingly, the client can use the intelligent financial counseling robot to handle the financial counseling business quite conveniently.
Owner:朱晓丹

Voice data automatic labeling method and system for voice recognition

The invention discloses an automatic voice data labeling method and system for voice recognition, and particularly relates to the field of voice recognition. The system comprises a mute detection module, a volume screening module, a length screening module, a voice recognition module, a recognition result judgment module and a manual proofreading module, the mute detection module splits each voiceinto a plurality of voice segments through a mute detection algorithm, and the volume screening module is used for screening out voices meeting the requirements through a volume threshold value and removing voices not meeting the requirements. The invention discloses a combined system of multiple modules. According to the system, speech preprocessing and speech recognition are carried out, by a public cloud mode, recognition result judgment manual proofreading are carried out, voice data annotation is constructed, after multiple times of iteration of the processes, a new corpus is continuously trained, high-quality corpus data is obtained, manpower is reduced, the voice data annotation quality is improved, and the problems that the manual annotation period is long, the cost is high and the efficiency is low are solved.
Owner:WEIFANG MEDICAL UNIV

Intelligent robot

The invention provides an intelligent robot. The robot includes a robot body. The robot body is provided with a speech system, an image system, a database processing system and a display screen. The database processing system is respectively connected with the speech system and the image system. The image system is used for collecting and processing a surrounding scene to obtain a three-dimensional scene. The speech system is used for collecting and recognizing a speech, which is uttered by a user, to obtain a speech instruction. The database processing system is used for storing a real scene,and invoking the three-dimensional scene according to the speech instruction. The display screen is used for displaying the invoked three-dimensional scene. The robot has the advantage of realizing user speech recognition and environment scene acquisition.
Owner:TIANJIN HUIZHI IOT TECH CO LTD

Chinese voice interaction non-inductive control system based on Raspberry Pi edge calculation and method thereof

The invention discloses a Chinese voice interaction non-inductive control system based on Raspberry Pi edge calculation and a method thereof, and the system comprises an edge end, a mobile end, an external control module, and an edge calculation detection and scheduling module. According to the invention, the sizes of a speech recognition model and a speech synthesis model are small; the edge calculation and offline work can be realized, the equipment is directly deployed in mobile terminal equipment without depending on a network, the functions of voice synthesis and voice recognition can berealized under an offline condition, the problems of voice recognition and interaction functions under severe conditions such as unsmooth network or attack are solved, and the function of Chinese voice recognition interaction under a severe environment is realized.
Owner:ENG UNIV OF THE CHINESE PEOPLES ARMED POLICE FORCE

Intelligent speech recognition recovery circuit and recovery method

The invention discloses an intelligent speech recognition recovery circuit and a recovery method, the recovery circuit comprises a speech control module, an audio processing module and an output module, the speech control module sends speech data to the audio processing module and receives the speech data; the audio processing module receives the speech data sent by the speech control module and further processes the speech data; and the output module receives and outputs the speech data further processed by the audio processing module. Preferably, the recovery circuit further comprises a sound effect algorithm processing module which is used for receiving the speech data sent by the speech control module, carrying out sound effect processing and returning the processed speech data to thespeech control module. According to the intelligent speech recognition and recovery circuit and the recovery method provided by the invention, the speech recognition function of a non-intelligent speech recognition product and an intelligent speech recognition product can be realized by saving the sampling circuit and the ADC conversion circuit.
Owner:SOUNDAI TECH CO LTD

Conference summary generation method and device, electronic equipment and storage medium

The invention discloses a conference summary generation method and device, electronic equipment and a storage medium. The method comprises the following steps: extracting a spectrogram of conference voice data; determining a first probability value between a signal feature of the conference voice data and a phoneme template according to the spectrogram by using an acoustic model of a preset intelligent decoding engine to obtain a phoneme feature corresponding to the signal feature, and determining a second probability value between the phoneme feature and a character template by using a language model of the preset intelligent decoding engine to obtain a phoneme feature corresponding to the phoneme feature; and decoding the conference voice data by using a decoder of a preset intelligent decoding engine according to the first probability value and the second probability value to obtain conference text data, thereby realizing end-to-end voice recognition without directly extracting voice features, and improving the voice recognition efficiency and accuracy in a complex scene. And finally, error correction operation is performed on the conference text data to generate a conference summary, so that the accuracy of a final result is further ensured.
Owner:广西中科曙光云计算有限公司 +1

Voice recognition method applied to user terminal and terminal equipment

The invention is suitable for the technical field of communication, and provides a voice recognition method applied to a user terminal and terminal equipment. The method includes the steps that voiceinformation input by a user is obtained, and corresponding character information is recognized according to the voice information; an interface type of a current display interface is determined and includes a list interface and a playing interface; when the current display interface is the list interface or the playing interface, the character information is matched with a regular expression correspnidng to the interface type of the current display interface. Semantic analysis is conducted at the user terminal, and will not be conducted in a semantic server, the process of semantic analysis can be simplified, and the time for semantic analysis is shortened; according to the determined regular expression selectively corresponding to the current display interface, regular expressions for matching in semantic analysis of the user terminal can be reduced, matching time is accordingly shortened, the time for semantic analysis of the user terminal is further shortened, the efficiency of recognizing voice is improved, and user experience is promoted.
Owner:TCL CORPORATION

Voice recognition method facing specific crowd

The invention discloses a voice recognition method facing a specific crowd. The method comprises the following steps of: first, sampling a voice signal and converting the voice signal to a digital signal from an analogue signal; then, pre-weighting, windowing, en-framing and performing front-end processing of endpoint detection on the digital voice signal; later on, performing feature extraction on the voice signal by adopting discrete wavelet transform; and finally, performing voice recognition on the feature-extracted voice signal by adopting a discrete hidden Markov model after training a sample. In the processes of performing the front-end processing and the feature extraction on the voice signal, spectrum features and pronunciation characteristics of different target crowds are fullytaken into consideration and the process of extracting voice information is optimized, so that a processing process and an information extracting process can be simplified; and therefore, recognitionprecision is ensured, simultaneously calculation amount and information storage capacity in the recognition process are greatly reduced, and the voice recognition on an embedded platform is realized.
Owner:HANGZHOU PINGPONG INTELLIGENT TECH CO LTD

Multifunctional intelligent wireless charging operation method controlled by computer

The invention relates to a multifunctional intelligent wireless charging operation method controlled by a computer. A charging device, a transmitting coil device, a power storage device, a monitoringdevice, an audio device, a central control device, a storage device, a communication device and an interaction device are included. The monitoring device collects the state data of a power bank in real time and stores the state data to the storage device, and the current working state is adjusted according to an optimal resource regulation and control model after the state data is processed by thecentral control device. According to the present invention, the intelligent voice control and audio playing functions are realized, and the interaction device can check the state of the multifunctional intelligent wireless power bank and control the multifunctional intelligent wireless power bank remotely.
Owner:潘小胜

Paint spraying robot voice recognition method based on multi-scale enhanced BiLSTM model

The invention discloses a paint spraying robot voice recognition method based on a multi-scale enhanced BiLSTM model. The method comprises the following steps: 1) acquiring common spraying sound instructions by using a signal acquisition system, wherein NI-9234 is selected as a data acquisition card; 2) repeatedly adding Gaussian white noise to the collected audio signal for 100 times, generating a noisy signal, solving a corresponding Mel spectrum sequence, and then solving an average sequence of 100 Mel spectrum sequences; 3) performing feature extraction on the average Mel spectrum sequence by using a multi-scale convolution filter, and then performing further mining on the extracted features by using a BiLSTM model to obtain corresponding output; 4) splicing outputs of the BiLSTM model together, then inputting the spliced outputs to a full connection layer and a Softmax layer for processing, and finally realizing speech recognition in combination with a CTC algorithm. and 5) embedding the model obtained through training in the steps 1) to 4) into a spraying robot, so that corresponding spraying tasks are intelligently achieved. According to the model, the intelligent voice recognition function of the spraying robot can be realized, and the model has very high practical application value.
Owner:JINLING INST OF TECH

Method and system for realizing voice age and/or gender recognition service, and medium

The invention relates to the field of voice recognition and particularly relates to a method, a system and a device for realizing voice age and / or gender recognition service and a medium, and aims to solve technical problems of remote accurate calling and simple and convenient deployment of an existing voice age and / or gender recognition model. Therefore, a terminal calls a server through a serialized voice age / gender identification request under a predefined GRPC framework, and the server identifies the age / gender through a set age / gender voice identification service; the corresponding voice age / gender recognition deep neural network model is accurately selected to decode and determine the age and / or gender information of the target object, and the age and / or gender information is returned to the terminal. Due to the fact that the age and / or gender service mode and the remote calling architecture are set, the corresponding model is called after the type of the model is determined, calling is more accurate and does not need to depend on a fixed frame, the method is more flexible, expandability is high, the resource utilization rate is high, concurrency is high, and meanwhile iterative updating of the algorithm model is facilitated.
Owner:GUANGZHOU YUNCONG INFORMATION TECH CO LTD

Electronic lock

The invention discloses an electronic lock, the structure of which comprises a LOGO slot, a protective cover, a fingerprint identifier, a fingerprint hole, an electronic host, a controller, a recording hole, a locker, a door handle, a power connector, a casing, a data line, The display chip, the audio chip, the control chip, and the locker have a rectangular structure and are arranged at the rear end of the electronic host by buckling. The front end of the electronic host is connected to the bottom of the fingerprint reader by fitting. Set up the buckle of the controller. The beneficial effects of the present invention are: the controller provided in the utility model analyzes the sound through the audio chip, and when the sound is confirmed to be the user, the unlocking control is carried out through the control chip, realizing the voice recognition function, greatly improving the functionality, and using more convenient.
Owner:GUANGZHOU DANJUE COMM TECH CO LTD

Voice recognition method, device, equipment, system and storage medium

The invention provides a voice recognition method, device, equipment and system and a storage medium. The method comprises steps of transmitting a voice recognition request to a server, and enabling the voice recognition request to comprise a to-be-recognized voice; obtaining a decoding recognition result of the to-be-recognized voice sent by the server; determining a speech recognition result corresponding to the speech to be recognized according to a pre-constructed hot word bank and the decoding recognition result; wherein a hot word database stores hot words corresponding to a user who sends the to-be-recognized voice. According to the method, personalized user voice recognition can be realized, and safety of the personalized information of the user can be ensured.
Owner:UNIV OF SCI & TECH OF CHINA +1

Speech recognition method and system based on comparative predictive coding

The invention discloses a voice recognition method and system based on comparative predictive coding, and belongs to the technical field of voiceprint recognition, and the method is characterized in that the method comprises the following steps: S1, collecting A voice files of each voice category, carrying out the preprocessing of each voice file, and obtaining PCM-coded voice time sequence data; s2, constructing a pairing data set of the voice time sequence data; s3, constructing a paired fragment data set; s4, constructing an artificial neural network; s5, training a speech recognition network formed by the first converter, the second converter and the one-dimensional convolutional neural network; and S6, voice recognition is carried out through the voice recognition network. According to the method, a large amount of insufficient voice data acquired by a background is fully utilized, the voice data are regarded as time sequence data, end-to-end conversion is directly realized, and the extraction of voice time sequence data characteristics is not needed.
Owner:北京信工博特智能科技有限公司

Speech recognition method and device, equipment and computer readable storage medium

PendingCN112951210AImprove accuracyImprove user experience satisfactionSpeech recognitionWord listSpeech sound
The invention relates to the technical field of speech recognition, and discloses a speech recognition method and device, equipment and a computer readable storage medium. According to the invention, if first voice information is monitored to be received, acoustic feature extraction is carried out on the first voice information to obtain first acoustic feature information, and then a decoder is utilized to decode the first acoustic feature information to obtain a decoding recognition result; and mode matching is performed on word strings in the decoding recognition result and word strings in a preset proper noun vocabulary to obtain a matching result, and the matching result is output to realize voice recognition. The problem that the accuracy of speech recognition is poor in the prior art is solved.
Owner:WOHO INNOVATION PLATFORM (SHENZHEN) CO LTD

Anti-theft wireless USB flash disk with voice recognition function

InactiveCN104392743ARealize photoelectric chargingImplement speech recognitionDigital storagePersonalizationComputer module
The invention discloses auxiliary computer hardware and particularly provides an anti-theft wireless USB flash disk with a voice recognition function. The anti-theft wireless USB flash disk comprises a storage end and a receiving end, wherein the storage end comprises a storage end shell, a flash memory unit, a storage end microprocessor, a storage end wireless receiving-transmitting module and a rechargeable battery; the receiving end comprises a receiving end shell, a receiving end USB interface, a receiving end wireless receiving-transmitting module and a receiving end microprocessor; and the anti-theft wireless USB flash disk further comprises a voice recognition device, a photocell, a photoelectric charger, a displacement sensor, an audible and visual alarm and a photosensitive sensor. According to the anti-theft wireless USB flash disk, the voice recognition can be realized, and photoelectric charging, automatic formatting and leakage prevention after stealing, and abnormal photoelectric alarm also can be realized, so that the information safety can be brought to a user; and the anti-theft wireless USB flash disk is simple to operate and is very convenient to use, and the individualized requirements of markets can be met.
Owner:SICHUAN CINGHOO TECH

Playing method of sound control vehicular player

The invention discloses a playing method of a sound control vehicular player. First of all, a parallel table of player control driving signals and entry codes is established; then, a database of sound instruction sample characteristic values and the entry codes is established; next, a BP network model is established, the BP network model is enabled to input the sound instruction sample characteristic values and output the entry codes, data in the database of the sound instruction sample characteristic values and the entry codes is input into the BP network model, and the BP network model is trained until network errors of the BP network model are smaller than and equal to a preset error threshold; and during work, a voice instruction of a driver is acquired, after the voice instruction of the driver is preprocessed, a characteristic value is calculated and then input to the BP network model, a corresponding entry code is obtained, retrieval is carried out in the parallel table, and a corresponding driving control signal is found for driving a player to work. According to the invention, the application is convenient, the driver does not have to move for operating the player, and thus the driving safety is improved.
Owner:苏州南光电子科技有限公司

Device for transferring speech recognition to video

The invention discloses a transforming device from speech identification to video, which comprises: an identifying code establishing module which is used for establishing a corresponding identifying code according to the types of video resource when a media server is started; an audio stream receiving module which is connected with the identifying code establishing module and is used for establishing a connecting channel of the audio stream and receiving the audio stream after the media server receives the request of an application server; a speech identifying module which is connected with the audio stream receiving module and is used for identifying audio data and outputting the identified data to a transformation processing module; the transformation processing module which is connected with the speech identifying module and the identifying code establishing module and is used for transforming the received data of the speech identifying code and making comparison between the transformed data with an identifying code established by the identifying code establishing module, thus realizing video transformation; a video stream output module which is connected with the transformation processing module and used for outputting the transformed video stream to a terminal unit through network.
Owner:ZTE CORP
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products