IEICE Trans - Predictive Trellis-Coded Quantization of the Cepstral Coefficients for the Distributed Speech Recognition

Predictive Trellis-Coded Quantization of the Cepstral Coefficients for the Distributed Speech Recognition

Sangwon KANG
Joonseok LEE

Publication
IEICE TRANSACTIONS on Communications Vol.E90-B No.6 pp.1570-1572
Publication Date: 2007/06/01
Online ISSN: 1745-1345
DOI: 10.1093/ietcom/e90-b.6.1570
Print ISSN: 0916-8516
Type of Manuscript: LETTER
Category: Multimedia Systems for Communications
Keyword:
distributed speech recognition (DSR), cepstral coefficients, quantization, BC-TCQ,

Full Text: PDF(258.4KB)>>

Summary:
In this paper, we propose a predictive block-constrained trellis-coded quantization (BC-TCQ) to quantize cepstral coefficients for distributed speech recognition. For prediction of the cepstral coefficients, the first order auto-regressive (AR) predictor is used. To quantize the prediction error signal effectively, we use the BC-TCQ. The quantization is compared to the split vector quantizers used in the ETSI standard, and is shown to lower cepstral distance and bit rates.