Internet-accessible speech recognition technology

Author

Huang, K. ; Picone, J.

Author_Institution

Inst. for Signal & Infortnation Process., Mississippi State Univ., MS, USA

Volume

3

fYear

2002

fDate

4-7 Aug. 2002

Abstract

Speech recognition systems can be viewed as an application of complex pattern recognition and machine learning algorithms. The development of such a system is a time-consuming and infrastructure-intensive task. The Institute for Signal and Information Processing (ISIP) developed one of the first fully-functional public domain speech recognition systems. In this paper, we introduce the major components of this system which include: a digital signal processing front end that generates feature vectors from the speech signal, a hidden Markov Model (HMM) trainer which estimates acoustic model parameters and a hierarchical search decoder which implements an efficient time-synchronous Viterbi beam search.

Keywords

Internet; decoding; hidden Markov models; learning (artificial intelligence); parameter estimation; search problems; speech recognition; DSP front end; HMM trainer; Internet-accessible speech recognition technology; acoustic model parameters; digital signal processing front end; feature vectors; hidden Markov model trainer; hierarchical search decoder; machine learning algorithms; pattern recognition; time-synchronous Viterbi beam search; Digital signal processing; Hidden Markov models; Information processing; Internet; Machine learning algorithms; Pattern recognition; Signal generators; Signal processing; Speech processing; Speech recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 2002. MWSCAS-2002. The 2002 45th Midwest Symposium on

Print_ISBN

0-7803-7523-8

Type

conf

DOI

10.1109/MWSCAS.2002.1186973

Filename

1186973