Automatic Speech Recognition - Explore the Science & Experts | ideXlab

Scan Science and Technology

Contact Leading Edge Experts & Companies

Automatic Speech Recognition

The Experts below are selected from a list of 303 Experts worldwide ranked by ideXlab platform

Jean-paul Haton – One of the best experts on this subject based on the ideXlab platform.

  • Automatic Speech Recognition: A Review
    Enterprise Information Systems V, 2005
    Co-Authors: Jean-paul Haton

    Abstract:

    Automatic Speech Recognition (ASR) has been extensively studied during the past few decades. Most of present systems are based on statistical modeling, both at the acoustic and linguistic levels, not only for Recognition, but also for understanding. Speech Recognition in adverse conditions has recently received increased attention since noise resistance has become one of the major bottlenecks for practical use of Speech recognizers. After briefly recalling the basic principles of statistical approaches to ASR (especially in a Bayesian framework), we present the types of solutions that have been proposed so far in order to obtain good performance in real life conditions.

  • Neural networks for Automatic Speech Recognition: a review
    Speech Processing Recognition and Artificial Neural Networks, 1999
    Co-Authors: Jean-paul Haton

    Abstract:

    Most present Automatic Speech Recognition systems are based on stochastic models, especially hidden Markov models (HMMs). However, during the past ten years, several projects have been directed toward the use of a new class of models: the connectionist artificial neural networks (ANNs).

  • Fundamentals of Automatic Speech Recognition
    Robustness in Automatic Speech Recognition, 1996
    Co-Authors: Jean-claude Junqua, Jean-paul Haton

    Abstract:

    After summarizing the difficulties encountered in Automatic Speech Recognition (ASR), we briefly describe the main approaches to ASR and present a historical review. We proceed by introducing popular distance measures used to evaluate the differences between extracted parameters. Then, we focus on the main pattern Recognition approaches, namely dynamic programming algorithms, stochastic modeling, and neural networks. We conclude this chapter by reviewing speaker-dependent and speaker-independent Recognition along with common discriminant methods used to improve ASR of confusable words.

Ivo Ipšić – One of the best experts on this subject based on the ideXlab platform.

  • Croatian Large Vocabulary Automatic Speech Recognition
    Automatika — Journal for Control Measurement Electronics Computing and Communications, 2011
    Co-Authors: Sanda Martinčić-ipšić, Miran Pobar, Ivo Ipšić

    Abstract:

    This paper presents procedures used for development of a Croatian large vocabulary Automatic Speech Recognition system (LVASR). The proposed acoustic model is based on context-dependent triphone hidden Markov models and Croatian phonetic rules. Different acoustic and language models, developed using a large collection of Croatian Speech, are discussed and compared. The paper proposes the best feature vectors and acoustic modeling procedures using which lowest word error rates for Croatian Speech are achieved. In addition, Croatian language modeling procedures are evaluated and adopted for speaker independent spontaneous Speech Recognition. Presented experiments and results show that the proposed approach for Automatic Speech Recognition using context-dependent acoustic modeling based on Croatian phonetic rules and a parameter tying procedure can be used for efficient Croatian large vocabulary Speech Recognition with word error rates below 5%.

Sanda Martinčić-ipšić – One of the best experts on this subject based on the ideXlab platform.

  • Croatian Large Vocabulary Automatic Speech Recognition
    Automatika — Journal for Control Measurement Electronics Computing and Communications, 2011
    Co-Authors: Sanda Martinčić-ipšić, Miran Pobar, Ivo Ipšić

    Abstract:

    This paper presents procedures used for development of a Croatian large vocabulary Automatic Speech Recognition system (LVASR). The proposed acoustic model is based on context-dependent triphone hidden Markov models and Croatian phonetic rules. Different acoustic and language models, developed using a large collection of Croatian Speech, are discussed and compared. The paper proposes the best feature vectors and acoustic modeling procedures using which lowest word error rates for Croatian Speech are achieved. In addition, Croatian language modeling procedures are evaluated and adopted for speaker independent spontaneous Speech Recognition. Presented experiments and results show that the proposed approach for Automatic Speech Recognition using context-dependent acoustic modeling based on Croatian phonetic rules and a parameter tying procedure can be used for efficient Croatian large vocabulary Speech Recognition with word error rates below 5%.