Maximum likelihood linear transformations for HMM-based speech recognition

A rule of thumb for obtaining good generalization in systems trained by examples is that one should use the smallest system that will fit the data. Unfortunately, it usually is not obvious what size is best; a system that is too small will not be able to learn the data while one that is just big enough may learn very slowly and be very sensitive to initial conditions and learning parameters. This paper is a survey of neural network pruning algorithms. The approach taken by the methods described here is to train a network that is larger than necessary and then remove the parts that are not needed. >

Pruning algorithms-a survey

An intelligent automated assistant system engages with the user in an integrated, conversational manner using natural language dialog, and invokes external services when appropriate to obtain information or perform various actions. The system can be implemented using any of a number of different platforms, such as the web, email, smartphone, and the like, or any combination thereof. In one embodiment, the system is based on sets of interrelated domains and tasks, and employs additional functionally powered by external services with which the system can interact.

Intelligent Automated Assistant

Systems and methods are provided for receiving speech and non-speech communications of natural language questions and/or commands, transcribing the speech and non-speech communications to textual messages, and executing the questions and/or commands. The invention applies context, prior information, domain knowledge, and user specific profile data to achieve a natural environment for one or more users presenting questions or commands across multiple domains. The systems and methods creates, stores and uses extensive personal profile information for each user, thereby improving the reliability of determining the context of the speech and non-speech communications and presenting the expected results for a particular question or command.

Systems and methods for responding to natural language speech utterance

Deep learning is becoming a mainstream technology for speech recognition at industrial scale. In this paper, we provide an overview of the work by Microsoft speech researchers since 2009 in this area, focusing on more recent advances which shed light to the basic capabilities and limitations of the current deep learning technology. We organize this overview along the feature-domain and model-domain dimensions according to the conventional approach to analyzing speech systems. Selected experimental results, including speech recognition and related applications such as spoken dialogue and language modeling, are presented to demonstrate and analyze the strengths and weaknesses of the techniques described in the paper. Potential improvement of these techniques and future research directions are discussed.

/pdf/recent-advances-in-deep-learning-for-speech-research-at-47expscg67.pdf

Recent advances in deep learning for speech research at Microsoft

Presents a maximum-likelihood (ML) stochastic matching approach to decrease the acoustic mismatch between a test utterance and a given set of speech models so as to reduce the recognition performance degradation caused by distortions in the test utterance and/or the model set. We assume that the speech signal is modeled by a set of subword hidden Markov models (HMM) /spl Lambda//sub x/. The mismatch between the observed test utterance Y and the models /spl Lambda//sub x/ can be reduced in two ways: 1) by an inverse distortion function F/sub /spl nu//(.) that maps Y into an utterance X that matches better with the models /spl Lambda//sub x/ and 2) by a model transformation function G/sub /spl eta//(.) that maps /spl Lambda//sub x/ to the transformed model /spl Lambda//sub x/ that matches better with the utterance Y. We assume the functional form of the transformations F/sub /spl nu//(.) or G/sub /spl eta//(.) and estimate the parameters /spl nu/ or /spl eta/ in a ML manner using the expectation-maximization (EM) algorithm. The choice of the form of F/sub /spl nu//(.) or G/sub /spl eta//(.) is based on prior knowledge of the nature of the acoustic mismatch. The stochastic matching algorithm operates only on the given test utterance and the given set of speech models, and no additional training data is required for the estimation of the mismatch prior to actual testing. Experimental results are presented to study the properties of the proposed algorithm and to verify the efficacy of the approach in improving the performance of a HMM-based continuous speech recognition system in the presence of mismatch due to different transducers and transmission channels.

A maximum-likelihood approach to stochastic matching for robust speech recognition

A computerized pronunciation system is provided for generating pronunciations for words and storing the pronunciations in a pronunciation dictionary. The system includes a word list including at least one word; transcribed acoustic data including at least one waveform for the word and transcribed text associated with the waveform; a pronunciation-learning module configured to accept as input the word list and the transcribed acoustic data, the pronunciation-learning module including: sets of initial pronunciations of the word, a scoring module configured score pronunciations and to generate phone probabilities, and a set of alternate pronunciations of the word, wherein the set of alternate pronunciations include a highest-scoring set of initial pronunciations with a highest-scoring substitute phone substituted for a lowest-probability phone; and a pronunciation dictionary configured to receive the highest-scoring set of initial pronunciations and the set of alternate pronunciations.

Method and system for learning linguistically valid word pronunciations from acoustic data

Combining the recognition outputs of multiple systems using methods such as ROVER or its extensions gives improved performance. However, previous approaches have been somewhat adhoc. We present BAYCOM, a Bayesian decision-theoretic approach to model combination that is optimal under given assumptions. We present recognition experiments showing that BAYCOM gives significant improvements over previous combination methods. In addition, we show that BAYCOM provides a confidence feature that gives very large improvements over previous methods for utterance rejection.

Bayesian model combination (BAYCOM) for improved recognition

In a speaker-independent, large-vocabulary continuous speech recognition systems, recognition accuracy varies considerably from speaker to speaker, and performance may be significantly degraded for outlier speakers such as nonnative talkers. In this paper, we explore supervised speaker adaptation and normalization in the MLP component of a hybrid hidden Markov model/ multilayer perceptron version of SRI's DECIPHERTM speech recognition system. Normalization is implemented through an additional transformation network that preprocesses the cepstral input to the MLP. Adaptation is accomplished through incremental retraining of the MLP weights on adaptation data. Our approach combines both adaptation and normalization in a single, consistent manner, works with limited adaptation data, and is text-independent. We show significant improvement in recognition accuracy.

Connectionist speaker normalization and adaptation.

In previous work, we showed how to constrain the estimation of continuous mixture-density hidden Markov models (HMMs) when the amount of adaptation data is small. We used maximum-likelihood (ML) transformation-based approaches and Bayesian techniques to achieve near native performance when testing nonnative speakers of the recognizer language. In this paper, we study various ML-based techniques and compare experimental results on data sets with recordings from nonnative and native speakers of American English. We divide the transformation-based techniques into two groups. In feature space techniques, we hypothesize an underlying transformation in the feature-space that results in a transformation of the HMM parameters. In model-space techniques, we hypothesize a direct transformation of the HMM parameters. In the experimental section we show how the combination of the best ML and Bayesian adaptation techniques result in significant improvements in recognition accuracy. All the experiments were carried out with SRI's DECIPHER TM speech recognition system [1][2].

Ananth Sankar

Papers

A maximum-likelihood approach to stochastic matching for robust speech recognition

Method and system for learning linguistically valid word pronunciations from acoustic data

Bayesian model combination (BAYCOM) for improved recognition

Connectionist speaker normalization and adaptation.

A comparative study of speaker adaptation techniques.