1481 results about "Acoustic model" patented technology

An automatic speech recognition system includes: a sound source localization module for localizing a sound direction of a speaker based on the acoustic signals detected by the plurality of microphones; a sound source separation module for separating a speech signal of the speaker from the acoustic signals according to the sound direction; an acoustic model memory which stores direction-dependent acoustic models that are adjusted to a plurality of directions at intervals; an acoustic model composition module which composes an acoustic model adjusted to the sound direction, which is localized by the sound source localization module, based on the direction-dependent acoustic models, the acoustic model composition module storing the acoustic model in the acoustic model memory; and a speech recognition module which recognizes the features extracted by a feature extractor as character information using the acoustic model composed by the acoustic model composition module.

The error rate of a pronunciation guesser that guesses the phonetic spelling of words used in speech recognition is improved by causing its training to weigh letter-to-phoneme mappings used as data in such training as a function of the frequency of the words in which such mappings occur. Preferably the ratio of the weight to word frequency increases as word frequencies decreases. Acoustic phoneme models for use in speech recognition with phonetic spellings generated by a pronunciation guesser that makes errors are trained against word models whose phonetic spellings have been generated by a pronunciation guesser that makes similar errors. As a result, the acoustic models represent blends of phoneme sounds that reflect the spelling errors made by the pronunciation guessers. Speech recognition enabled systems are made by storing in them both a pronunciation guesser and a corresponding set of such blended acoustic models.

A system for substantially automating transcription services for multiple voice users including a manual transcription station, a speech recognition program and a routing program. The system establishes a profile for each of the voice users containing a training status which is selected from the group of enrollment, training, automated and stop automation. When the system receives a voice dictation file from a current voice user based on the training status the system routes the voice dictation file to a manual transcription station and the speech recognition program. A human transcriptionist creates transcribed files for each received voice dictation files. The speech recognition program automatically creates a written text for each received voice dictation file if the training status of the current user is training or automated. A verbatim file is manually established if the training status of the current user is enrollment or training and the speech recognition program is trained with an acoustic model for the current user using the verbatim file and the voice dictation file if the training status of the current user is enrollment or training. The transcribed file is returned to the current user if the training status of the current user is enrollment or training or the written text is returned if the training status of the current user is automated. An apparatus and method is also disclosed for simplifying the manual establishment of the verbatim file. A method for substantially automating transcription services is also disclosed.

A domain-based speech recognition method and apparatus, the method including: performing speech recognition by using a first language model and generating a first recognition result including a plurality of first recognition sentences; selecting a plurality of candidate domains, by using a word included in each of the first recognition sentences and having a confidence score equal to or higher than a predetermined threshold, as a domain keyword; performing speech recognition with the first recognition result, by using an acoustic model specific to each of the candidate domains and a second language model and generating a plurality of second recognition sentences; and selecting at least one or more final recognition sentence from the first recognition sentences and the second recognition sentences. According to this method and apparatus, the effect of a domain extraction error by misrecognition of a word on selection of a final recognition result can be minimized.

Provided is a method and apparatus for searching music based on speech recognition. By calculating search scores with respect to a speech input using an acoustic model, calculating preferences in music using a user preference model, reflecting the preferences in the search scores, and extracting a music list according to the search scores in which the preferences are reflected, a personal expression of a search result using speech recognition can be achieved, and an error or imperfection of a speech recognition result can be compensated for.

According to one embodiment, a method, apparatus for synthesizing speech, and a method for training acoustic model used in speech synthesis is provided. The method for synthesizing speech may include determining data generated by text analysis as fuzzy heteronym data, performing fuzzy heteronym prediction on the fuzzy heteronym data to output a plurality of candidate pronunciations of the fuzzy heteronym data and probabilities thereof, generating fuzzy context feature labels based on the plurality of candidate pronunciations and probabilities thereof, determining model parameters for the fuzzy context feature labels based on acoustic model with fuzzy decision tree, generating speech parameters from the model parameters, and synthesizing the speech parameters via synthesizer as speech.

A speech recognition system includes the following: a feature calculating unit; a sound level calculating unit that calculates an input sound level in each frame; a decoding unit that matches the feature of each frame with an acoustic model and a linguistic model, and outputs a recognized word sequence; a start-point detector that determines a start frame of a speech section based on a reference value; an end-point detector that determines an end frame of the speech section based on a reference value; and a reference value updating unit that updates the reference value in accordance with variations in the input sound level. The start-point detector updates the start frame every time the reference value is updated. The decoding unit starts matching before being notified of the end frame and corrects the matching results every time it is notified of the start frame. The speech recognition system can suppress a delay in response time while performing speech recognition based on a proper speech section.

Certain disclosed methods and systems perform multiple different types of message recognition using a shared language model. Message recognition of a first type is performed responsive to a first type of message input (e.g., speech), to provide text data in accordance with both the shared language model and a first model specific to the first type of message recognition (e.g., an acoustic model). Message recognition of a second type is performed responsive to a second type of message input (e.g., handwriting), to provide text data in accordance with both the shared language model and a second model specific to the second type of message recognition (e.g., a model that determines basic units of handwriting conveyed by freehand input). Accuracy of both such message recognizers can be improved by user correction of misrecognition by either one of them. Numerous other methods and systems are also disclosed.

The present invention provides a computerized method and apparatus for automatically generating from a first speech recognizer a second speech recognizer which can be adapted to a specific domain. The first speech recognizer can include a first acoustic model with a first decision network and corresponding first phonetic contexts. The first acoustic model can be used as a starting point for the adaptation process. A second acoustic model with a second decision network and corresponding second phonetic contexts for the second speech recognizer can be generated by re-estimating the first decision network and the corresponding first phonetic contexts based on domain-specific training data.

A mobile terminal (100) which is controllable by spoken utterances like proper names or command words is described. The mobile terminal (100) comprises an interface (200) for receiving from a network server (300) acoustic models for automatic speech recognition and an automatic speech recognizer (110) for recognizing the spoken utterances based on the received acoustic models. The invention further relates to a network server (300) for mobile terminals (100) which are controllable by spoken utterances and to a method for obtaining acoustic models for a mobile terminal (100) controllable by spoken utterances.

A system and method of testing and tuning a speech recognition system by providing pronunciations to the speech recognizer. First a text document is provided to the system and converted into a sequence of phonemes representative of the words in the text. The phonemes are then converted to model units, such as Hidden Markov Models. From the models a probability is obtained for each model or state, and feature vectors are determined. The feature vector matching the most probable vector for each state is selected for each model. These ideal feature vectors are provided to the speech recognizer, and processed. The end result is compared with the original text, and modifications to the system can be made based on the output text.

A system is provided for training an acoustic model for use in speech recognition. In particular, such a system may be used to perform training based on a spoken audio stream and a non-literal transcript of the spoken audio stream. Such a system may identify text in the non-literal transcript which represents concepts having multiple spoken forms. The system may attempt to identify the actual spoken form in the audio stream which produced the corresponding text in the non-literal transcript, and thereby produce a revised transcript which more accurately represents the spoken audio stream. The revised, and more accurate, transcript may be used to train the acoustic model using discriminative training techniques, thereby producing a better acoustic model than that which would be produced using conventional techniques, which perform training based directly on the original non-literal transcript.

A method and system for endpoint automatic detection of audio record is provided. The method comprises the following steps: acquiring a audio record text and affirming the text endpoint acoustic model for the audio record text; starting acquiring the audio record data of each frame in turn from the audio record start frame in the audio record data; affirming the characteristics acoustic model of the decoding optimal path for the acquired current frame of the audio record data; comparing the characteristics acoustic model of the decoding optimal path acquired from the current frame of the audio record data with the endpoint acoustic model to determine if they are the same; if yes, updating a mute duration threshold with a second time threshold, wherein the second time threshold is less than a first time threshold. This method can improve the recognizing efficiency of the audio record endpoint.

Systems and processes for automatic accent detection are provided. In accordance with one example, a method includes, at an electronic device with one or more processors and memory, receiving a user input, determining a first similarity between a representation of the user input and a first acoustic model of a plurality of acoustic models, and determining a second similarity between the representation of the user input and a second acoustic model of the plurality of acoustic models. The method further includes determining whether the first similarity is greater than the second similarity. In accordance with a determination that the first similarity is greater than the second similarity, the first acoustic model may be selected; and in accordance with a determination that the first similarity is not greater than the second similarity, the second acoustic model may be selected.

A query may be applied against search engines that respectively return a set of search results relating to various items discovered in the searched data sets. However, presenting numerous and varied search results may be difficult on mobile devices with small displays and limited computational resources. Instead, search results may be associated with search domains representing various information types (e.g., contacts, public figures, places, projects, movies, music, and books) and presented by grouping search results with associated query domains, e.g., in a tabbed user interface. The query may be received through an input device associated with a particular input domain, and may be transitioned to the query domain of a particular search engine (e.g., by recognizing phonemes of a voice query using an acoustic model; matching phonemes with query terms according to a pronunciation model; and generating a recognition result according to a vocabulary of an n-gram language model.)

An input system for mobile search and a method therefor are provided. The input system includes an input module receiving a code input for a specific term and a voice input corresponding thereto, a database including a glossary and an acoustic model, wherein the glossary includes a plurality of terms and a sequence list, and each of the terms has a search weight based on an order of the sequence list, a process module selecting a first number of candidate terms from the glossary according to the code input by using an input algorithm and obtaining a second number of candidate terms by using a speech recognition algorithm to compare the voice input with the first number of candidate terms via the acoustic model, wherein the second number of candidate terms are listed in a particular order based on their respective search weights, and an output module showing the second number of candidate terms in the particular order for selecting the specific term therefrom.

To provide a data process unit and data process unit control program which are suitable for generating acoustic models for unspecified speakers taking distribution of diversifying feature parameters into consideration under such specific conditions as the type of speaker, speech lexicons, speech styles, and speech environment and which are suitable for providing acoustic models intended for unspecified speakers and adapted to speech of a specific person.

A data process unit 1 comprises a data classification section 1a, data storing section 1b, pattern model generating section 1c, data control section 1d, mathematical distance calculating section 1e, pattern model converting section 1f, pattern model display section 1g, region dividing section 1h, division changing section 1i, region selecting section 1j, and specific pattern model generating section 1k.

The invention discloses an implementation method and application of a voice awakening module. The implementation method comprises the following steps of: voice input (1), voice awakening algorithm (2) and awakening actuation (3), wherein the voice awakening algorithm (2) is implemented through the following main steps of: acoustic feature extraction (4), awakening word detection (5), awakening word confirmation (6), construction of an awakening word detection network (7), training of an acoustic model (8) and construction of an awakening word confirming network (9) and the like. The invention has the advantages that even under a noisy environment, no matter whether the music is played, the voice awakening function can be started by the voice awakening word, and the recognition awakening effect is good; and the implementation method can be planted onto an ARM or DSP universal process for operation and is applied in the fields related to vehicle mounting and household appliances.

A system for substantially automating transcription services for multiple users (10, 11, 12) including a manual transcription station (50), speech recognition program (40) and a routing program (200). A uniquely identified voice dictation file is generated from a user and—based on the training status—routes the voice dictation file to a manual transcription station and speech recognition program. A human transcriptionist creates transcribed files for each voice dictation file. The speech recognition program creates written text for each dictation file if the training status is training or automated. If the training status of the current user is enrollment or training, a verbatim file is manually established and the speech recognition program is trained with an acoustic model using the verbatim and voice dictation files. The transcribed file is returned to the user if the training status is enrollment or training or written text is returned if the status is automated.