Abstract: A method, program product, and system for speech recognition, the method comprising in one embodiment pruning a hypothesis based on a first criteria; storing information about the pruned hypothesis; and reactivating the pruned hypothesis if a second criterion is met. In an embodiment, the first criteria may be that another hypothesis has a better score at that time by some predetermined amount. In an embodiment, the stored information may comprise at least one of a score for the pruned hypothesis, an identification of the hypothesis that caused the pruning and the frame in which the pruning took place. In a further embodiment, the reactivating step may use at least some of the stored information about the pruned hypothesis in performing the reactivation and the second criteria may be that a revised score for the hypothesis that caused the pruning is worse by some predetermined amount from an original expected score calculated for that hypothesis.

Abstract: The invention disclosed herein concerns a method of converting speech to text using a hierarchy of contextual models. The hierarchy of contextual models can be statistically smoothed into a language model. The method can include processing text with a plurality of contextual models. Each one of the plurality of contextual models can correspond to a node in a hierarchy of the plurality of contextual models. Also included can be identifying at least one of the contextual models relating to the text and processing subsequent user spoken utterances with the identified at least one contextual model.

Abstract: Methods and apparatus for facilitating speaker recognition, wherein, from target data that is provided relating to a target speaker and background data that is provided relating to at least one background speaker, a set of cohort data is selected from the background data that has at least one proximate characteristic with respect to the target data. The target data and the cohort data are then combined in a manner to produce at least one new cohort model for use in subsequent speaker verification. Similar methods and apparatus are contemplated for non-voice-based applications, such as verification through fingerprints.

Abstract: A method and apparatus for enhancing coding efficiency by reducing illegal or other undesirable packet generation while encoding a signal. The probability of generating illegal or other undesirable packets while encoding a signal is reduced by first analyzing a history of the frequency of codebook values selected while quantizing speech parameters. Codebook entries are then reordered so that the index/indices that create illegal or other undesirable packets contain the least frequently used entry/entries. Reordering multiple codebooks for various parameters further reduces the probability that an illegal or other undesirable packet will be created during signal encoding. The method and apparatus may be applied to reduce the probability of generating illegal null traffic channel data packets while encoding eighth rate speech.

Abstract: A method and system that combines voice recognition engines and resolves differences between the results of individual voice recognition engines using a mapping function. Speaker independent voice recognition engines and speaker-dependent voice recognition engines are combined. Hidden Markov Model (HMM) engines and Dynamic Time Warping (DTW) engines are combined.

Abstract: A method is provided for real-time speaker change detection and speaker tracking in a speech signal. The method is a “coarse-to-refine” process, which consists of two stages: pre-segmentation and refinement. In the pre-segmentation process, the covariance of a feature vector of each segment of speech is built initially. A distance is determined based on the covariance of the current segment and a previous segment; and the distance is used to determine if there is a potential speaker change between these two segments. If there is no speaker change, the model of current identified speaker model is updated by incorporating data of the current segment. Otherwise, if there is a speaker change, a refinement process is utilized to confirm the potential speaker change point.

Abstract: A speech recognition system for recognizing an input voice of a narrow frequency band. The speech recognition system includes: a frequency band converting unit for converting the input voice of the narrow frequency band into a pseudo voice of a wide frequency band which covers an entirety of the narrow frequency band and which is wider than the narrow frequency band.

Abstract: A method and apparatus is provided for matching a first sequence of patterns representative of a first signal with a second sequence of patterns representative of a second signal. The system uses a plurality of different pruning thresholds (th) to control the propagation of paths which represent possible matchings between a sequence of second signal patterns and a sequence of first signal patterns ending at the current first signal pattern. In particular, the pruning threshold used for a given path during the processing of a current first signal pattern depends upon the position, within the sequence of patterns representing the second signal, of the second signal pattern which is at the end of the given path.

Abstract: A speech recognition support method in a system to retrieve a map in response to a user's input speech. The user's speech is recognized and a recognition result is obtained. If the recognition result represents a point on the map, a distance between the point and a base point on the map is calculated. The distance is decided to be above a threshold or not. If the distance is above the threshold, an inquiry to confirm whether the recognition result is correct is output to the user.

Abstract: A speech-recognizing apparatus for recognizing input speech comprises, an analysis unit for computing a characteristic vector for each of frames of the input speech, a correction-value storage unit for storing a correction distance in advance, a vector-to-vector-distance-computing unit for computing a vector-to-vector distance between the characteristic vector and the phoneme characteristic vector, an average-value-computing unit for computing an average value of vector-to-vector distances for one of the frames, a correction unit for computing a corrected vector-to-vector distance as a value of an expression of (the vector-to-vector distance-the average value+the correction distance), and a recognition unit for cumulating corrected vector-to-vector distances into a cumulative vector-to-vector distance and comparing the cumulative vector-to-vector distance with the word standard pattern in order to recognize the input speech.

Abstract: A method includes (i) measuring first distances between (a) vectors belonging to a set of vectors that represent an utterance and (b) vectors belonging to a set of vectors that represent a template, the measuring being done in accordance with a first order of the utterance vectors a first order of the template vectors, and (ii) measuring second distances between (a) individual vectors belonging to the set of vectors that represent the utterance and (b) individual vectors belonging to the set of vectors that represent the template, the measuring being done in accordance with a second order of the utterance vectors and a second order of the template vectors, and (iii) in which the first template vector order and the second template vector order are different and/or the first utterance vector order and the second utterance vector order are different.

Abstract: A method is provided for improving pattern matching in a speech recognition system having a plurality of acoustic models. The improved method includes: receiving continuous speech input; generating a sequence of acoustic feature vectors that represent temporal and spectral behavior of the speech input; loading a first group of acoustic feature vectors from the sequence of acoustic feature vectors into a memory workspace accessible to a processor; loading an acoustic model from the plurality of acoustic models into the memory workspace; and determining a similarity measure for each acoustic feature vector of the first group of acoustic feature vectors in relation to the acoustic model. Prior to retrieving another group of acoustic feature vectors, similarity measures are computed for the first group of acoustic feature vectors in relation to each of the acoustic models employed by the speech recognition system.

Abstract: A recognizing target vocabulary comparing unit calculates a compared likelihood of a recognizing target vocabulary, i.e., a compared likelihood of a registered vocabulary, by using the time series of the amount of characteristics of an input speech. An environment adaptive noise model comparing unit calculates a compared likelihood of a noise model adaptive to a noise environment, i.e., a compared likelihood of environmental noise. A rejection determining unit compares the likelihood of the registered vocabulary with the likelihood of the environmental noise, and determines whether or not the input speech is the noise. When it is determined that the input speech is the noise, a noise model adapting unit adaptively updates an environment adaptive noise model by using the input speech. Thus, the environment adaptive noise model matches to a real environment and the rejection determination can be performed for a noise input with high accuracy.

Abstract: A phonetic distance calculation method for similarity comparison between phonetic transcriptions of foreign words. A system manager defines character element transformation patterns occurrable between phonetic transcriptions derived from the same foreign language. A system generates new phonetic transcriptions according to the defined character element transformation patterns and assigns a demerit mark to each of the generated phonetic transcriptions according to a phonetic distance. A minimum phonetic distance between each of the generated phonetic transcriptions and a given phonetic transcription is calculated on the basis of a minimum edit distance calculation method. Any one of the generated phonetic transcriptions with a smallest one of the calculated minimum phonetic distances is determined to be most similar to the given phonetic transcription.

Abstract: A system and method for building compressed biometric models and performing biometric identification using such models. The use of the compressed biometric models results in a significant decrease in the storage requirements for biometric models in conventional biometric systems. A given number of L reference biometric models are built. The L reference models are randomly divided into M subsets. During user enrollment, distance measurements between a temporary biometric model and each of the reference models in the M subsets are computed.

Abstract: A voice recognition rejection scheme for capturing an utterance includes the steps accepting the utterance, applying an N-best algorithm to the utterance, or rejecting the utterance. The utterance is accepted if a first predefined relationship exists between one or more closest comparison results for the utterance with respect to a stored word and one or more differences between the one or more closest comparison results and one or more other comparison results between the utterance and one or more other stored words. An N-best algorithm is applied to the utterance if a second predefined relationship exists between the one or more closest comparison results and the one or more differences between the one or more closest comparison results and the one or more other comparison results.

Abstract: A method for concatenative speech synthesis includes a processing stage that selects segments based on their symbolic labeling in an efficient graph-based search, which uses a finite-state transducer formalism. This graph-based search uses a representation of concatenation constraints and costs that does not necessarily grow with the size of the source corpus thereby limiting the increase in computation required for the search as the size of the source corpus increases. In one application of this method, multiple alternative segment sequences are generated and a best segment sequence is then be selected using characteristics that depend on specific signal characteristics of the segments.

Abstract: A method for generating a hierarchical speaker model tree. In an illustrative embodiment, a speaker model is generated for each of a number of speakers from which speech samples have been obtained. Each speaker model contains a collection of distributions of audio feature data derived from the speech sample of the associated speaker. The hierarchical speaker model tree is created by merging similar speaker models on a layer by layer basis. Each time two or more speaker models are merged, a corresponding parent speaker model is created in the next higher layer of the tree. The tree is useful in applications such as speaker verification and speaker identification. A speaker verification method is disclosed in which a claimed ID from a claimant is received, where the claimed ID represents a speaker corresponding to a particular one of the speaker models. A cohort set of similar speaker models associated with the particular speaker model is established.

Abstract: Systems and methods for classifying natural language (NL) sentences using a combination of NL algorithms or techniques is disclosed. Each NL algorithm or technique may identify a different similarity trait between two or more sentences, and each may help compare the meaning of the sentences. By combining the various similarity factors, preferably by various weighting factors, a distance metric can be computed. The distance metric provides a measure of the overall similarity between sentences, and can be used to assign a sentences to an appropriate sentence category.

Abstract: The distance computation represents a central, constantly recurrent task in sample and speech recognition. It is used in speech recognition as a degree of similarity between a part of a speech utterance and a speech reference. In picture processing and sample recognition, it is used for data compression. The distance computation requires the longest computation time so that a reduction of the computation time results in a considerable efficiency improvement. A reduction of the computation time is achieved by the integration of the distance computation in a memory module in which particularly the reference data are stored. Due to this integration, the other components of the overall system are relieved of this constantly recurrent task and are available for more complex processes in this period of time. This integration makes the distance computation essentially shorter because the communication between memory sections and computation unit takes place directly without utilizing a busy system.

Abstract: The speech recognition device, which can realize speech recognition with a small-scaled circuit, has been disclosed. The speech recognition device comprises the similarity circuit, which receives speech input signals and puts out characteristics based on the self-organizing algorithm, and the matrix circuit that performs the matrix operations of the output signal, wherein: the similarity circuit comprises a circuit that calculates distances between plural multi-dimensional input vectors and the pattern vectors prepared in advance, calculates a value corresponding to one dimension using a pair of neuron MOSFETS, and forms a voltage signal in accordance with the degree of similarity by summing up the current that flows in each neuron MOSFET; and the matrix circuit, in which capacitors corresponding to weighting operations are arranged in matrix, receives a voltage signal in accordance with the degree of similarity and outputs what is most similar, to the patterns prepared in advance.

Abstract: A system and method for synthesizing a facial image, compares a speech frame from an incoming speech signal with acoustic features stored within visually similar entries in an audio-visual codebook to produce a set of weights. The audio-visual codebook also stores visual features corresponding to the acoustic features. A composite visual feature is generated as a weighted sum of the corresponding visual features, from which the facial image is synthesized. The audio-visual codebook may include multiple samples of the acoustic and visual features for each entry, which corresponds to a sequence of one or more phonemes.

Abstract: A method and a system is disclosed that provide means to enable individuals with speech, language and reading based communication disabilities, due to a temporal processing problem, to improve their temporal processing abilities as well as their communication abilities. The method and system include provisions to elongate portions of phonemes that have brief and/or rapidly changing acoustic spectra, such as occur in the stop consonants b and d in the phonemes /ba/ and /da/, as well as reduce the duration of the steady state portion of the syllable. In addition, some emphasis is added to the rapidly changing segments of these phonemes. Additionally, the disclosure includes method for and computer software to modify fluent speech to make the modified speech better recognizable by communicatively impaired individuals. Finally, the disclosure includes method for and computer software to train temporal processing abilities, specifically speed and precision of temporal integration, sequencing and serial memory.

Abstract: A method and a system is disclosed that provide means to enable individuals with speech, language and reading based communication disabilities, due to a temporal processing problem, to improve their temporal processing abilities as well as their communication abilities. The method and system include provisions to elongate portions of phonemes that have brief and/or rapidly changing acoustic spectra, such as occur in the stop consonants b and d in the phonemes /ba/ and /da/, as well as reduce the duration of the steady state portion of the syllable. In addition, some emphasis is added to the rapidly changing segments of these phonemes. Additionally, the disclosure includes method for and computer software to modify fluent speech to make the modified speech better recognizable by communicatively impaired individuals. Finally, the disclosure includes method for and computer software to train temporal processing abilities, specifically speed and precision of temporal integration, sequencing and serial memory.

Abstract: A method and a system is disclosed that provide means to enable individuals with speech, language and reading based communication disabilities, due to a temporal processing problem, to improve their temporal processing abilities as well as their communication abilities. The method and system include provisions to elongate portions of phonemes that have brief and/or rapidly changing acoustic spectra, such as occur in the stop consonants b and d in the phonemes /ba/ and /da/, as well as reduce the duration of the steady state portion of the syllable. In addition, some emphasis is added to the rapidly changing segments of these phonemes. Additionally, the disclosure includes method for and computer software to modify fluent speech to make the modified speech better recognizable by communicatively impaired individuals. Finally, the disclosure includes method for and computer software to train temporal processing abilities, specifically speed and precision of temporal integration, sequencing and serial memory.

Abstract: A method and a system is disclosed that provide means to enable individuals with speech, language and reading based communication disabilities, due to a temporal processing problem, to improve their temporal processing abilities as well as their communication abilities. The method and system include provisions to elongate portions of phonemes that have brief and/or rapidly changing acoustic spectra, such as occur in the stop consonants b and d in the phonemes /ba/ and /da/, as well as reduce the duration of the steady state portion of the syllable. In addition, some emphasis is added to the rapidly changing segments of these phonemes. Additionally, the disclosure includes method for and computer software to modify fluent speech to make the modified speech better recognizable by communicatively impaired individuals. Finally, the disclosure includes method for and computer software to train temporal processing abilities, specifically speed and precision of temporal integration, sequencing and serial memory.

Abstract: A method and a system is disclosed that provide means to enable individuals with speech, language and reading based communication disabilities, due to a temporal processing problem, to improve their temporal processing abilities as well as their communication abilities. The method and system include provisions to elongate portions of phonemes that have brief and/or rapidly changing acoustic spectra, such as occur in the stop consonants b and d in the phonemes /ba/ and /da/, as well as reduce the duration of the steady state portion of the syllable. In addition, some emphasis is added to the rapidly changing segments of these phonemes. Additionally, the disclosure includes method for and computer software to modify fluent speech to make the modified speech better recognizable by communicatively impaired individuals. Finally, the disclosure includes method for and computer software to train temporal processing abilities, specifically speed and precision of temporal integration, sequencing and serial memory.

Abstract: A method and a system is disclosed that provide means to enable individuals with speech, language and reading based communication disabilities, due to a temporal processing problem, to improve their temporal processing abilities as well as their communication abilities. The method and system include provisions to elongate portions of phonemes that have brief and/or rapidly changing acoustic spectra, such as occur in the stop consonants b and d in-the phonemes /ba/ and /da/, as well as reduce the duration of the steady state portion of the syllable. In addition, some emphasis is added to the rapidly changing segments of these phonemes. Additionally, the disclosure includes method for and computer software to modify fluent speech to make the modified speech better recognizable by communicatively impaired individuals. Finally, the disclosure includes method for and computer software to train temporal processing abilities, specifically speed and precision of temporal integration, sequencing and serial memory.

Abstract: A method and a system is disclosed that provide means to enable individuals with speech, language and reading based communication disabilities, due to a temporal processing problem, to improve their temporal processing abilities as well as their communication abilities. The method and system include provisions to elongate portions of phonemes that have brief and/or rapidly changing acoustic spectra, such as occur in the stop consonants b and d in the phonemes /ba/ and /da/, as well as reduce the duration of the steady state portion of the syllable. In addition, some emphasis is added to the rapidly changing segments of these phonemes. Additionally, the disclosure includes method for and computer software to modify fluent speech to make the modified speech better recognizable by communicatively impaired individuals. Finally, the disclosure includes method for and computer software to train temporal processing abilities, specifically speed and precision of temporal integration, sequencing and serial memory.

Abstract: An apparatus for selecting a cohort model for use in a speaker verification system includes a model generator (108) for determining a target speaker model (114) from a speech sample collected from the target speaker (106). A cohort selector (110) determines a similarity value between each of a number of predetermined existing speaker models from a model pool (112) and the target speaker model (114) and a dissimilarity value between each of the existing speaker models and any previously selected cohort models (116). An existing speaker model which is most similar to the target speaker model, but most dissimilar to previously chosen cohort models, is then chosen as another cohort model for the target speaker.

Abstract: A method of recognizing a speaker of an input speech according to the distance between an input speech pattern, obtained by converting the input speech to a feature parameter series, and a reference pattern preliminarily registered as feature parameter series for each speaker is provided. Contents of the input and reference speech patterns is obtained by recognition. An identical section, in which the contents of the input and reference speech patterns are identical is determined. The distance between the input and reference speech patterns in the calculated identical content section is determined. The speaker of the input speech is recognized on the basis of the determined distance.

Abstract: The distance computation represents a central, constantly recurrent task in sample and speech recognition. It is used in speech recognition as a degree of similarity between a part of a speech utterance and a speech reference. In picture processing and sample recognition, it is used for data compression (MPEG). The distance computation requires the longest computation time so that a reduction of the computation time results in a considerable efficiency improvement. A reduction of the computation time is achieved by the integration of the distance computation in a memory module (1) in which particularly the reference data are stored. Due to this integration, the other components (2, 3, 4) of the overall system are relieved of this constantly recurrent task and are available for more complex processes in this period of time.

Abstract: In a computerized method a continuous signal is segmented in order to determine statistically stationary units of the signal. The continuous signal is sampled at periodic intervals to produce a timed sequence of digital samples. Fixed numbers of adjacent digital samples are grouped into a plurality of disjoint sets or frames. A statistical distance between adjacent frames is determined. The adjacent sets are merged into a larger set of samples or cluster if the statistical distance is less than a predetermined threshold. In an iterative process, the statistical distance between the adjacent sets are determined, and as long as the distance is less than the predetermined threshold, the sets are iteratively merged to segment the signal into statistically stationary units.

Abstract: A string of acoustic feature parameters of each of recognition-desired words and a string of acoustic feature parameters of each of reception words are registered in advance. When an uttered word is received, a string of acoustic feature parameters is extracted from the uttered word, the acoustic feature parameters of the uttered word is compared with the string of acoustic feature parameters of each recognition-desired word, and a recognition-desired word recognition score indicating a similarity degree between the uttered word and each recognition-desired word is calculated. Also, a reception word recognition score indicating a similarity degree between the uttered word and each reception word is calculated.

Abstract: An pattern dissimilarity calculator according to the present invention, which calculates a pattern dissimilarity between a first and second sequence feature pattern using either the DP matching approach or the HMM (Hidden Marcov Model) approach, comprises cumulative distance calculator 1 for calculating the distance between frame i of said first sequence feature pattern and each of frames of said second sequence feature pattern, and obtaining a current cumulative distance by adding to an cumulative distance obtained in terms of frame i−1, which is decoded in cumulative decoder 4; and cumulative distance encoder 2 for encoding the cumulative distance calculated by the cumulative distance calculator 1. The cumulative distance decoder 4 decodes cumulative distances encoded by the encoding means.

Abstract: Speech recognition systems and methods consistent with the present invention process input speech signals organized into a series of frames. The input speech signal is decimated to select K frames out of every L frames of the input speech signal according to a decimation rate K/L. A first set of model distances is then calculated for each of the K selected frames of the input speech signal, and a Hidden Markov Model (HMM) topology of a first set of models is reduced according to the decimation rate K/L. The system then selects a reduced set of model distances from the computed first set of model distances according to the reduced HMM topology and selects a first plurality of candidate choices for recognition according to the reduced set of model distances. A second set of model distances is computed, using a second set of models, for a second plurality of candidate choices, wherein the second plurality of candidate choices correspond to at least a subset of the first plurality of candidate choices.

Abstract: A method for computing a distance between collections of distributions or finite mixture models of features. Data is processed so as to define at least first and second collections of distributions of features. For each distribution of the first collection, the distance to each distribution of the second collection is measured to determine which distribution of the second collection is the closest (most similar). The same procedure is performed for the distributions of the second collection. Based on the closest distance measures, a final distance is computed representing the distance between the first and second collections. This final distance may be a weighted sum of the closest distances. The distance measure may be used in a number of applications such as [speaker classification,] speaker recognition and audio segmentation.

Abstract: A speech recognition method and apparatus in which a speech section is sliced by the unit of a word by spotting and candidate words are selected. Next, in a second stage, matching is conducted by the unit of a phoneme. Consequently, selection of the candidate words and slicing of the speech section can be performed concurrently. Furthermore, narrowing of the candidate words is facilitated. Furthermore, since reference phoneme patterns under a plurality of environments are prepared, recognition of an input speech under a larger number of conditions is possible using a smaller amount of data when compared with the case in which reference word patterns under a plurality of environments are prepared.

Abstract: A method and apparatus for matching a first sequence of patterns representative of a first signal with a second sequence of patterns representative of a second signal using a dynamic programming matching technique is described. The second signal patterns which are at the end of a dynamic programming path for a current first signal pattern are listed in an active list 201. The dynamic programming paths are propagated by processing the second signal patterns on the active list, and a new active list 205 is generated for the succeeding input pattern. In order to propagate each path, the system determines how many second signal patterns lie within an overlap region in which a comparison has to be made, and processes each path in dependence upon the determined amount of overlap.

Abstract: A method is provided for reducing search complexity in a speech recognition system having a fast match, a detailed match, and a language model. Based on at least one predetermined variable, the fast match is optionally employed to generate candidate words and acoustic scores corresponding to the candidate words. The language model is employed to generate language model scores. The acoustic scores are combined with the language model scores and the combined scores are ranked to determine top ranking candidate words to be later processed by the detailed match, when the fast match is employed. The detailed match is employed to generate detailed match scores for the top ranking candidate words.

Abstract: Speech models are constructed and trained upon the speech of known client speakers (and also impostor speakers, in the case of speaker verification). Parameters from these models are concatenated to define supervectors and a linear transformation upon these supervectors results in a dimensionality reduction yielding a low-dimensional space called eigenspace. The training speakers are then represented as points or distributions in eigenspace. Thereafter, new speech data from the test speaker is placed into eigenspace through a similar linear transformation and the proximity in eigenspace of the test speaker to the training speakers serves to authenticate or identify the test speaker.

Abstract: In a speech recognition method and system a spoken word to be recognized is broken down into input vectors (K2a), ambient noise is evaluated (K1), a recognized word is chosen from a dictionary having associated reference vectors which are separated from the input vectors by a shortest distance (K3), and the recognized word is validated by a comparison of this distance with a threshold value which is derived as a function of the result of the evaluation of ambient noise. The ambient noise evaluation may be carried out at an instant when the speaker is silent, i.e. either before or after the speaker speaks the word to be recognized.

Abstract: A high-speed speech recognition method with a high recognition rate, utilizing speaker models, includes the steps of executing an acoustic process on the input speech, calculating a coarse output probability utilizing an unspecified speaker model, and calculating a fine output probability utilizing an unspecified speaker model and clustered speaker models, for the states estimated, by the result of coarse calculation, to contribute to the results of recognition. Candidates of recognition are then extracted by a common language search based on the obtained result, and a fine language search is conducted on the thus extracted candidates to determine the result of recognition.

Abstract: Speech processing is obtained that, given a probabilistic mapping between static speech sounds and pseudo-articulator positions, allows sequences of speech sounds to be mapped to smooth sequences of pseudo-articulator positions. In addition, a method for learning a probabilistic mapping between static speech sounds and pseudo-articulator position is described. The method for learning the mapping between static speech sounds and pseudo-articulator position uses a set of training data composed only of speech sounds. The said speech processing can be applied to various speech analysis tasks, including speech recognition, speaker recognition, speech coding, speech synthesis, and voice mimicry.

Abstract: One embodiment of a speech recognition system is organized with speech input signal preprocessing and feature extraction followed by a fuzzy matrix quantizer (FMQ). Frames of the speech input signal are represented by a vector .function. of line spectral pair frequencies and are fuzzy matrix quantized to respective a vector .function. entries in a codebook of the FMQ. A distance measure between .function. and .function., d(.function.,.function.), is defined as ##EQU1## where the constants .alpha..sub.1, a.sub.2, .beta..sub.1 and .beta..sub.2 are set to substantially minimize quantization error, and e.sub.i is the error power spectrum of the speech input signal and a predicted speech input signal at the ith line spectral pair frequency of the speech input signal. The speech recognition system may also include hidden Markov models and neural networks, such as a multilevel perceptron neural network, speech classifiers.

Abstract: One embodiment of a speech recognition system is organized with speech input signal preprocessing and feature extraction followed by a fuzzy matrix quantizer (FMQ). Frames of the speech input signal are represented in a matrix by a vectorf of line spectral pair frequencies and energy coefficients and are fuzzy matrix quantized to respective vector f entries of a matrix codeword in a codebook of the FMQ. The energy coefficients include the original energy and the first and second derivatives of the original energy which increase recognition accuracy by, for example, being generally distinctive speech input signal parameters and providing noise signal suppression especially when the noise signal has a relatively constant energy over at least two time frame intervals. To reduce data while maintaining sufficient resolution, the energy coefficients may be normalized and logarithmically represented. A distance measure between f and f, d(f, f), is defined as ##EQU1## where the constants .alpha..sub.1, .alpha..sub.

Abstract: For coding human speech for subsequent audio reproduction thereof, a plurality of speech segments is derived from speech received, and systematically stored in a data base for later concatenated readout. After the deriving, respective speech segments are fragmented into temporally consecutive source frames, similar source frames as governed by a predetermined similarity measure thereamongst that is based on an underlying parameter set are joined, and joined source frames are collectively mapped onto a single storage frame. Respective segments are stored as containing sequenced referrals to storage frames for therefrom reconstituting the segment in question.

Abstract: Methods and systems consistent with the present invention enroll a candidate phrase uttered by a user in a dictionary having at least one previously enrolled phrase. The system receives utterances of the candidate phrase and determines whether the first utterance is confusingly similar to a previously enrolled phrase and whether they are consistent with each other. The system then enrolls the candidate phrase in the dictionary according to these determinations.

Abstract: A speech recognition system, in which partial reference patterns, and cumulative similarities of these patterns, are stored in a temporary pattern memory. The partial reference patterns are to be used as subjects of a similarity computation with an input speech pattern that has its feature quantities extracted by a speech analyzing unit. A counting unit counts partial reference patterns having corresponding cumulative similarities that are higher than a threshold value stored in a threshold memory. A threshold computing unit computes a threshold of pruning from a correspondence relation between the number of partial reference patterns that have corresponding cumulative similarities that exceed the threshold, and the threshold. A similarity computing unit computes a similarity, with respect to the feature quantities, of partial reference patterns with corresponding cumulative similarities that are greater than the threshold of pruning.

Abstract: A speech recognition apparatus has an analysis section that outputs features of input speech as a time sequence of feature vectors defined for discrete time points corresponding to a processed speech frame. Reference paradigm utterances are converted into a time sequence of standard (reference) feature vectors. The possible continuous variation of standard feature vectors at each point in time is expressed by a line segment, or set of line segments, connecting the feature vectors for the two end points of the "movable" range within which the feature can change, rather than using a larger set of reference vectors as in a conventional multitemplate approach to speech recognition. For example, the continuous range of possible background noise levels in input speech defines a line segment connecting the two feature vectors at the two SNR value limits.