A method and apparatus are provided for performing speech recognition using a dynamic vocabulary. Results from a preliminary speech recognition pass can be used to update or refine a language model in order to improve the accuracy of search results and to simplify subsequent recognition passes. This iterative process greatly reduces the number of alternative hypotheses produced during each speech recognition pass, as well as the time required to process subsequent passes, making the speech recognition process faster, more efficient and more accurate. The iterative process is characterized by the use of results from one or more data set queries, where the keys used to query the data set, as well as the queries themselves, are constructed in a manner that produces more effective language models for use in subsequent attempts at decoding a given speech signal.

Method and apparatus for speech recognition using a dynamic vocabulary

A set of speaker dependent models or adapted models is trained upon a comparatively large number of training speakers, one model per speaker, and model parameters are extracted in a predefined order to construct a set of supervectors, one per speaker. Dimensionality reduction is then performed on the set of supervectors to generate a set of eigenvectors that define an eigenvoice space. If desired, the number of vectors may be reduced to achieve data compression. Thereafter, a new speaker provides adaptation data from which a supervector is constructed by constraining this supervector to be in the eigenvoice space based on a maximum likelihood estimation. The resulting coefficients in the eigenspace of this new speaker may then be used to construct a new set of model parameters from which an adapted model is constructed for that speaker. The adapted model may then be further adapted via MAP, MLLR, MLED or the like. The eigenvoice technique may be applied to MLLR transformation matrices or the like; Bayesian estimation performed in eigenspace uses prior knowledge about speaker space density to refine the estimate about the location of a new speaker in eigenspace.

Dimensionality reduction for speaker normalization and speaker and environment adaptation using eigenvoice techniques

Acoustic features (109, 111) are extracted from input speech (107) and are compared (113) against pre-stored models (117). The result is used to make a judgement of the user's pronunciation (115).

Method and system for automatic text-independent grading of pronunciation for language instruction

The invention is a system and method for automatic acoustic speaker adaptation in an automatic speech recognition assisted transcription system. Partial transcripts of audio files are generated by a transcriptionist. A topic language model is generated from the partial transcripts. The topic language model is interpolated with a general language model. Automatic speech recognition is performed on the audio files by a speech recognition engine using a speaker independent acoustic model and the interpolated language model to generate semi-literal transcripts of the audio files. The semi-literal transcripts are then used with the corresponding audio files to generate a speaker dependent acoustic model in an acoustic adaptation engine.

Systems and methods for automatic acoustic speaker adaptation in computer-assisted transcription systems

A method of gender dependent speech recognition includes the steps of identifying phone state models common to both genders, identifying gender specific phone state models, identifying a gender of a speaker and recognizing acoustic data from the speaker. A method of constructing a gender-dependent speech recognition model includes the steps of providing training data of a known gender, aligning the training data, tagging the training data with a gender to create gender-tagged data, determining a gender question at a node to determine gender dependence of the gender-tagged data, determining a phonetic context question at the node to determine phonetic context dependence of the gender-tagged data, determining a highest value of an evaluation function between the gender dependence and the phonetic context dependence to determine which dependence is a dominant dependence, splitting the data of the dominant dependence into child nodes according to likelihood criteria, comparing the highest value with a threshold value to determine if additional splitting is necessary, repeating theses steps for each child node until the highest value is below the threshold value and counting the nodes having gender dependence to determine an overall gender dependence level. A gender-dependent speech recognition system includes an input device for inputting speech to a preprocessor. The preprocessor converts the speech into acoustic data, and a processor for identifies gender-dependent phone state models and phone state modes common to both genders. The phone state models are stored in a memory device wherein the processor recognizes the speech in accordance with the phone state models.

Speech recognition models combining gender-dependent and gender-independent phone states and using phonetic-context-dependence

In accordance with the invention, a speech recognizer is provided which uses a computationally-feasible method for constructing a set of Hidden Markov Models (HMMs) for speech recognition that utilize a partial and optimal degree of mixture tying. With partially-tied HMMs, improved recognition accuracy of a large vocabulary word corpus as compared to systems that use fully-tied HMMs is achieved with less computational overhead than with a fully untied system. The computationally-feasible technique comprises the steps of determining a cluster of HMM states that share Gaussian components which are close together, developing a subset codebook for those clusters, and recalculating the Gaussians in the codebook to best estimate the clustered states.

Murveit Hy

Papers

Method and apparatus for speech recognition using optimized partial mixture tying of HMM state functions

Verfahren und vorrichtung zur spracherkennung mittels optimierter partieller buendelung von wahrscheinlichkeitsmischungen