Discriminative training for discrete HMM of a fixed-point DSP Mandarin digits recognition system

Author

He, Qiang ; Liu, Jia ; Liu, Runsheng

Author_Institution

Dept. of Electr. Eng., Tsinghua Univ., Beijing, China

Volume

1

fYear

2002

fDate

26-30 Aug. 2002

Firstpage

532

Abstract

Voice dialing technology is widely used in mobile phones, but most of them are based on speaker-dependent speech recognition, and cannot support dialing by saying the digits directly. Digit dialing requires speaker-independent speech recognition technology. The paper introduces a 16-bit fixed-point DSP based Mandarin digit recognition system based on discrete HMM, and discriminative training for the parameters. The parameters of the DHMM are reduced and only the truncated output probability matrices are used. A minimum classification error rate method is used to adjust the matrices discriminatively to improve the recognition accuracy further. In an experiment, a 14.4% error rate reduction is achieved.

Keywords

error statistics; hidden Markov models; learning (artificial intelligence); speech recognition; speech-based user interfaces; Mandarin digit recognition; classification error rate; discrete HMM; discriminative training; fixed-point DSP; mobile phones; speaker-dependent speech recognition; speaker-independent speech recognition; truncated output probability matrices; voice dialing; Books; Digital signal processing; Engines; Error analysis; Helium; Hidden Markov models; Mel frequency cepstral coefficient; Mobile handsets; Samarium; Speech recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing, 2002 6th International Conference on

Print_ISBN

0-7803-7488-6

Type

conf

DOI

10.1109/ICOSP.2002.1181110

Filename

1181110