Multiple description quantization for recognizing voice over packet networks

I. Te Lin*, Chun Feng Wu, Sin Horng Chen, Wen Whei Chang

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The practical design of multiple description vector quantizers for robust distributed speech recognition over packet networks is investigated. In the proposed system, speech parameters are quantized and mapped to multiple descriptions for transmission over independent channels. A new approach to the index assignment optimization is presented on the basis of a linear programming framework. Also, a fast local search algorithm is proposed to find the optimal index assignment without compromising the speech recognition accuracy. Experiments with random packet loss in a range of loss conditions are conducted on the Mandarin digit string recognition task. Simulation results indicate that the proposed multiple description scheme provides more robust performance than the ETSI standardized split vector quantization scheme with a single description.

Original languageEnglish
Title of host publicationFirst International Conference on Communications and Networking in China, ChinaCom '06
PublisherIEEE Computer Society
ISBN (Print)1424404630, 9781424404636
DOIs
StatePublished - 1 Jan 2006
Event1st International Conference on Communications and Networking in China, ChinaCom '06 - Beijing, China
Duration: 25 Oct 200627 Oct 2006

Publication series

NameFirst International Conference on Communications and Networking in China, ChinaCom '06

Conference

Conference1st International Conference on Communications and Networking in China, ChinaCom '06
CountryChina
CityBeijing
Period25/10/0627/10/06

Fingerprint Dive into the research topics of 'Multiple description quantization for recognizing voice over packet networks'. Together they form a unique fingerprint.

Cite this