120results about "Speech recognition" patented technology

The present invention relates to a spoken dialog system and method based on dual dialog management using a hierarchical dialog task library that may increase reutilization of dialog knowledge by constructing and packaging the dialog knowledge based on a task unit having a hierarchical structure, and may construct and process the dialog knowledge using a dialog plan scheme about relationship therebetween by classifying the dialog knowledge based on a task unit to make design of a dialog service convenient, which is different from an existing spoken dialog system in which it is difficult to reuse dialog knowledge since a large amount of construction costs and time is required.

A voice recognition method for a single sentence including a multi-instruction in an interactive voice user interface, method includes steps of detecting a connection ending by analyzing the morphemes of a single sentence on which voice recognition has been performed, separating the single sentence into a plurality of passages based on the connection ending, detecting a multi-connection ending by analyzing the connection ending and extracting instructions by specifically analyzing passages including the multi-connection ending and outputting a multi-instruction included in the single sentence by combining the instructions extracted in the step of extracting instructions. In accordance with the present invention, consumer usability can be significantly increased because a multi-operation intention can be checked in one sentence.

A method of configuring a speech recognition unit in a vehicle. The method includes receiving a noise error from the speech recognition unit responsive to a user voice command and reducing a confidence threshold for an appropriate grammar set responsive to the received noise error.

A method and device for voiceprint recognition, include: establishing a first-level Deep Neural Network (DNN) model based on unlabeled speech data, the unlabeled speech data containing no speaker labels and the first-level DNN model specifying a plurality of basic voiceprint features for the unlabeled speech data; obtaining a plurality of high-level voiceprint features by tuning the first-level DNN model based on labeled speech data, the labeled speech data containing speech samples with respective speaker labels, and the tuning producing a second-level DNN model specifying the plurality of high-level voiceprint features; based on the second-level DNN model, registering a respective high-level voiceprint feature sequence for a user based on a registration speech sample received from the user; and performing speaker verification for the user based on the respective high-level voiceprint feature sequence registered for the user.

A method of constructing a text message on a mobile communications device, the method involving: storing a plurality of text phrases; for each of the text phrases, storing a representation that is derived from that text phrase; receiving a spoken phrase from a user; from the received spoken phrase generating an acoustic representation thereof; based on the acoustic representation, searching among the stored representations to identify a stored text phrase that best matches the spoken phrase; and inserting into an electronic document the text phrase that is identified from searching.

The invention discloses a conference record generation method based on a telephone conference and a device. The method comprises steps that voice content acquired by each conference terminal is acquired; the voice content is converted into text content; conference record is generated according to the text content, and the conference record is stored and/or is sent to a designated address. Through the method, the voice content recorded by each conference terminal is automatically converted into the text content by utilizing the voice identification technology, the conference record is generated according to the text content, so the conference record of the telephone conference is automatically generated, tediousness in manual conference recording can be avoided, operation efficiency is improved, and a telephone conference system is made to be more intelligent.

The invention provides an intelligent-robot-oriented dialog system data processing method and device. The method comprises the following steps: 1) based on an obtained interaction topic sequence, extracting a target topic from the interaction topic sequence according to a preset rule; 2) determining a corresponding output module according to the attribute of the target topic, and based on a preset knowledge graph, generating correlation information according to the attribute of the target topic; and 3) generating feedback information corresponding to the target topic according to the correlation information and the output module. Through the method, the dialog system can initiate some related topics actively from time to time for questions of users to allow the users to think the dialog system can understand their dialogue information truly, so that man-machine interaction is allowed to be carried on, and user experience and user viscosity of the dialog system are improved.

The invention relates to a speech interactive training system and a speech interactive training method. The system comprises a user selection module, a speech interactive training module, a user feedback module, a speech evaluation module and a result feedback module, wherein the user selection module is used for acquiring training contents selected by a user; the speech interactive training module is used for displaying the training contents to the user in a multimode guiding mode to guide the user to perform a speech training; the user feedback module is used for collecting a fed-back speech and a lip video corresponding to the speech; the speech evaluation module is used for receiving the speech fed back by the user and the lip video corresponding to the speech, and automatically evaluating the speech training of the user and giving an evaluation result; and the result feedback module is used for feeding the evaluation result back to the user so that the user can correct and adjust the speech training. The speech interactive training system is used for automatically evaluating the speech training of the user, giving the evaluation result and feeding the evaluation result back to the user, and then the user finds out the level of the personal speech training according to the evaluation result and corrects and adjusts the personal speech training to further improve the speech level, so the rehabitation training effect of a speech impediment is greatly enhanced.

The invention relates to a voice input method and system. The method includes: recording a voice and at the same time segmenting the input voice into voice segments and generating a text for each voice segment; and displaying the text of each voice segment in order and correcting the text of each voice segment in order according to a selection of a user. The voice input method and system enable a voice identification result to be segmented automatically and paragraphed and then returned for a second confirmation of the user so that the user can record a voice while correcting and confirming a returned text.

A computer-implemented method for providing an objective evaluation to a customer service representative regarding his performance during an interaction with a customer may include receiving a digitized data stream corresponding to a spoken conversation between a customer and a representative; converting the data stream to a text stream; generating a representative transcript that includes the words from the text stream that are spoken by the representative; comparing the representative transcript with a plurality of positive words and a plurality of negative words; and generating a score that varies according to the occurrence of each word spoken by the representative that matches one of the positive words, and/or the occurrence of each word spoken by the representative that matches one of the negative words. Tone of voice, as well as response time, during the interaction may also be monitored and analyzed to adjust the score, or generate a separate score.

The invention provides a voice interaction method. The method comprises receiving a first voice message; converting the first voice message into a first text message; searching a first result corresponding to the first text message according to the first text message; displaying the first result; receiving a second voice message; converting the second voice message into a second text message; comparing the first text message with the second text message to obtain a third text message; performing searching according to the third text message and based on the first result to obtain a second result; and displaying the second result. The invention further provides a voice interaction device. According to the voice interaction method and the device, a voice input this time is analyzed through a recognition result of the former time recognized combined with a user voice, factors of usage habits, external scenes and the like are combined to perform intelligent recognition, and the recognition result of the first time is screened, so that the recognition efficiency is accurate, and the user experience is improved.

The invention discloses a video playing method and device, a terminal device, and a storage medium. The method includes the steps: extracting an audio from a video, and generating an audio file; converting the audio file into a file stream, and converting a file stream into a caption text by voice recognition, wherein the caption text includes a plurality of time stamps corresponding to the playing time of the audio; displaying the caption text on a playing interface of the video according to the time stamps; receiving a query instruction including a keyword, and querying the target time stampcorresponding to the keyword in the caption text, wherein the plurality of time stamps include the target time stamp; playing the audio and the video according to the target time stamp. The method can output and display the caption text at a corresponding position on the video efficiently and quickly, and can accurately locate the playing position on the time axis of the video, thereby greatly improving the user experience.

A speech segment determination device includes a frame division portion, a power spectrum calculation portion, a power spectrum operation portion, a spectral entropy calculation portion and a determination portion. The frame division portion divides an input signal in units of frames. The power spectrum calculation portion calculates, using an analysis length, a power spectrum of the input signal for each of the frames that have been divided. The power spectrum operation portion adds a value of the calculated power spectrum to a value of power spectrum in each of frequency bins. The spectral entropy calculation portion calculates spectral entropy using the power spectrum whose value has been increased. The determination portion determines, based on a value of the spectral entropy, whether the input signal is a signal in a speech segment.

The invention discloses a self-adaptive endpoint detection method and a self-adaptive endpoint detection system for isolate word speech recognition. The self-adaptive endpoint detection method for isolate word speech recognition comprises the following steps: a, a voice input step, wherein a voice signal containing an isolate word to be recognized is input; b, a voice preprocessing step, wherein the voice signal is subjected to amplitude translation and normalization and framing processing operation, and short time average energy and a short time average zero-crossing rate of each frame of voice are calculated; c, an isolate word endpoint rough detection step, wherein isolate word endpoints are roughly estimated through utilization of the short time average energy and the short time average zero-crossing rate of each frame of the voice signal and constraint on the shortest length of continuous voice frames before and after the end points, d, a detection threshold self-adaptive adjustment and accurate endpoint detection step, wherein through utilization of constraint on the smallest time duration and the largest time duration of the isolate word, the detection threshold is subjected to dynamic adjustment operation, the voice endpoints are subjected to front and back fine adjustment, and accurate isolate word endpoints are obtained; e, an isolate word endpoint output and isolate word voice recognition step, wherein the accurate isolate word endpoints are output and isolate word recognition is realized by using voice recognizing technologies.

Methods and apparatus of managing a communication system, wherein a decision regarding a level of activity at a first end is made based at least in part on the level of activity at the second end. In one embodiment, the energy level of a first-end audio signal is measured. The first end is declared voice-active if the first-end energy level is greater than or equal to a first threshold value. The first end is declared voice-inactive if the first-end energy level is less than the first threshold value. To determine the value of the first threshold value, the energy level of a second-end audio signal is measured. If the second-end energy level is greater than or equal to a second threshold value, the second end is declared voice-active, in which case the first threshold is maintained at a relatively high level. If the second-end energy level is less than the second threshold value, the second end is declared voice-inactive, in which case the first threshold is maintained at a relatively lower level.

A system is provided that includes an inspection instrument. The inspection instrument includes communications circuitry, configured to communicatively couple the inspection instrument with information services; an audio input device, configured to receive inspection audio; and a processor, configured to query the information services via sending a request to the information services based upon an analysis of keywords, phrases, or both interpreted from the inspection audio. A voice recognition system configured to analyze the inspection audio for the keywords, phrases, or both is also provided.

The application relates to a data processing technology, and provides a locker control method and apparatus, computer equipment, and a storage medium. The method includes steps: obtaining voice information detected by a voice detector; extracting target voice content from the voice information through a speech recognition model, and extracting a target voiceprint characteristic from the voice information through the voiceprint recognition model; inquiring preset voice content matched with the target voice content and a preset voiceprint characteristic matched with the target voiceprint characteristic; inquiring a locker identification corresponding to the preset voiceprint characteristic when the preset voice content and the preset voiceprint characteristic are inquired; and transmitting adoor-opening control instruction to an electromagnetic lock corresponding to the locker identification, wherein the door-opening control instruction is used for instructing the electromagnetic lock to open a locker corresponding to the locker identification. By employing the method, the security of the locker can be improved.

The invention provides a voice command processing method and a corresponding electronic device. The voice command processing method is applied to the electronic device, and an interface including display objects corresponding to operations are displayed on the display screen of the electronic device. The voice command processing method comprises the steps of: receiving a first input implemented by a user in a first manner; determining a range in the interface according to the first input, wherein the range relates to one or more display objects; providing the range to the user; processing each related display object in the range, and thus determining a keyword for indicating each display object, wherein the keywords of the display objects are different from one another; providing the keywords to the user; using the keywords as matching words for implementing voice matching, so that when inputting the keywords through voice, the user executes the corresponding operation of the display object corresponding to the keword. According to the voice command processing method and the electronic device, the user can input a voice command conveniently and rapidly.

The invention belongs to the technical field of mobile terminals, and discloses a voice taxi calling method comprising the steps that voice information of a user is detected in real time; when the mobile terminal responds to preset awakening information included in the voice information of the user under the standby state, a taxi calling software client side is awakened; and when the taxi calling software client side responds to destination information included in the voice information of the user, current position information of the mobile terminal is acquired, and the current position information and the destination information are transmitted to a taxi calling software server so that the taxi calling software server is enabled to start the taxi calling flow. The voice information of the user is identified, the awakening information and the destination information are acquired from the voice information of the user, the taxi calling software client side is awakened according to the awakening information and the current position information of the mobile terminal is acquired, and the destination information and the current position information are transmitted to the taxi calling software server so as to start the taxi calling flow. The taxi calling service can be realized by inputting the destination information for one time through the voice information.

The invention provides a voice identification method, an apparatus and a terminal thereof, and a computer readable storage medium. The voice identification method comprises the following steps of through a voice assistant, acquiring voice information of a user of a terminal; acquiring current emotion information of the user corresponding to the voice information; through the voice assistant, according to a current emotion of the user and the voice information, determining output information corresponding to the voice information; and outputting the output information. In the invention, a literal meaning of the user can be identified, emotion changes of the user can be perceived and voice identification accuracy is increased; and simultaneously, voice identification is not inflexible and is humanized so that a good user experience is brought.

A self-improving approximator for use in media editing is disclosed. The approximator estimates location in the media file/video data domain of a user-selected word or text unit in the text script transcription of the corresponding audio of the video data. During editing, the approximator calculates and displays the estimated time location of user-selected text to assist the user-editor in cross referencing between the beginning and ending of user-selected passage statements in the text script and the corresponding video/media data in a rough cut or subsequent media work. The approximator enables simultaneous editing of text and video/media by the selection of either source component. The approximator self improves its accuracy based on differentials calculated between tracked user adjustments to media-text associations and initial approximations (estimates).

The invention provides a voice auto-answer cloud server, a voice auto-answer system and a voice auto-answer method. The voice auto-answer cloud server comprises a communication unit, a recognizing and matching unit and a research and response unit. The communication unit is used for receiving a voice request sent by a user. The recognizing and matching unit conducts recognition and fuzzy matching on the voice request so as to confirm user needs which match with the voice request, wherein each user need in the fuzzy matching process matches with a plurality of voice requests. The search and response unit searches corresponding response information according to the user needs and sends the response information to the user through a communication module. The voice auto-answer cloud server, the voice auto-answer system and the voice auto-answer method are suitable for various different voice instructions and are convenient to use.

The invention discloses a multi-fundamental frequency extraction method and a multi-fundamental frequency extraction device based on empirical mode decomposition and a hidden Markov model. The method comprises steps: an auditory filter bank is used for filtering a speech signal, and framing is carried out on the signal after filtering; an auto-correlation function is calculated on each time frequency unit for an auditory spectrum; on the basis of an intrinsic mode function obtained through the empirical mode decomposition, the instantaneous frequency of each time frequency unit dominant sound source is calculated; on the basis of each instantaneous frequency, a frequency matching function is calculated; the frequency matching function is used for building the likelihood probability of each fundamental frequency state, and a corpus is used for counting the transition probability between each fundamental frequency state and a fundamental frequency value; and the likelihood probability of each fundamental frequency state is enhanced, the enhanced likelihood probability is combined with the corresponding transition probability, and the hidden Markov model is used for extracting a multi-fundamental frequency track of the speech signal.

The invention relates to a voice control technology, and in particular relates to a method for searching audio and video resources via voice, which is applied to a smart TV set. The method for searching audio and video resources via voice provided by the invention can be summarized into the following steps: transferring voice information data received by a voice receiving device to a data processing module; then converting the data to text information data; transferring the text information data to a data transmitting module; searching a server for audio and video resources through an audio and video search module according to the content of the text information data; and finally processing the searched resources,then feeding back the processed searched resources to a user through a display module for selection by the user, and playing the audio and video resources selected by the user. The method for searching audio and video resources via voice has the beneficial effect that the user only needs to send an instruction via voice to accomplish search of the audio and video resources, and the search coverage is large, content can be automatically played according to the resource format, and user operation can be greatly simplified. The method is especially applicable to the smart TV set.