Compensation of Speech Coding Distortion for Wireless Speech Recognition

Hong Kook KIM

Compensation of Speech Coding Distortion for Wireless Speech Recognition

Hong Kook KIM

Full Text Views

0

Share
Cite this

Summary :

In this paper, we perform some experiments to show that the quantization noise caused by low-bit-rate speech coding can be characterized as a white noise process. Then, the signal-to-quantization noise ratio of the decoded speech for a given bit-rate is estimated by observing the perceptual speech quality equivalent to the artificially generated noisy speech obtained by adding a white Gaussian noise source. This information is incorporated into the parameter tuning of a noise-robust compensation algorithm for speech recognition so that the compensation algorithm can be performed better under a range of the estimated SNRs. Finally, we apply the compensation algorithm to a connected digit string recognition system that utilizes speech signals decoded by the GSM adaptive multi-rate (AMR) speech coder. It is shown that the noise-robust compensation algorithm reduces word error rates by 15% or more at low bit-rate modes of the AMR speech coder.

Publication: IEICE TRANSACTIONS on Information Vol.E87-D No.6 pp.1596-1600

Publication Date: 2004/06/01

Publicized

Online ISSN

DOI

Type of Manuscript: LETTER

Category: Speech and Hearing

Cite this

Copy

Hong Kook KIM, "Compensation of Speech Coding Distortion for Wireless Speech Recognition" in IEICE TRANSACTIONS on Information, vol. E87-D, no. 6, pp. 1596-1600, June 2004, doi: .
Abstract: In this paper, we perform some experiments to show that the quantization noise caused by low-bit-rate speech coding can be characterized as a white noise process. Then, the signal-to-quantization noise ratio of the decoded speech for a given bit-rate is estimated by observing the perceptual speech quality equivalent to the artificially generated noisy speech obtained by adding a white Gaussian noise source. This information is incorporated into the parameter tuning of a noise-robust compensation algorithm for speech recognition so that the compensation algorithm can be performed better under a range of the estimated SNRs. Finally, we apply the compensation algorithm to a connected digit string recognition system that utilizes speech signals decoded by the GSM adaptive multi-rate (AMR) speech coder. It is shown that the noise-robust compensation algorithm reduces word error rates by 15% or more at low bit-rate modes of the AMR speech coder.
URL: https://globals.ieice.org/en_transactions/information/10.1587/e87-d_6_1596/_p

Copy

@ARTICLE{e87-d_6_1596,
author={Hong Kook KIM, },
journal={IEICE TRANSACTIONS on Information},
title={Compensation of Speech Coding Distortion for Wireless Speech Recognition},
year={2004},
volume={E87-D},
number={6},
pages={1596-1600},
abstract={In this paper, we perform some experiments to show that the quantization noise caused by low-bit-rate speech coding can be characterized as a white noise process. Then, the signal-to-quantization noise ratio of the decoded speech for a given bit-rate is estimated by observing the perceptual speech quality equivalent to the artificially generated noisy speech obtained by adding a white Gaussian noise source. This information is incorporated into the parameter tuning of a noise-robust compensation algorithm for speech recognition so that the compensation algorithm can be performed better under a range of the estimated SNRs. Finally, we apply the compensation algorithm to a connected digit string recognition system that utilizes speech signals decoded by the GSM adaptive multi-rate (AMR) speech coder. It is shown that the noise-robust compensation algorithm reduces word error rates by 15% or more at low bit-rate modes of the AMR speech coder.},
keywords={},
doi={},
ISSN={},
month={June},}

Copy

TY - JOUR
TI - Compensation of Speech Coding Distortion for Wireless Speech Recognition
T2 - IEICE TRANSACTIONS on Information
SP - 1596
EP - 1600
AU - Hong Kook KIM
PY - 2004
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E87-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 2004
AB - In this paper, we perform some experiments to show that the quantization noise caused by low-bit-rate speech coding can be characterized as a white noise process. Then, the signal-to-quantization noise ratio of the decoded speech for a given bit-rate is estimated by observing the perceptual speech quality equivalent to the artificially generated noisy speech obtained by adding a white Gaussian noise source. This information is incorporated into the parameter tuning of a noise-robust compensation algorithm for speech recognition so that the compensation algorithm can be performed better under a range of the estimated SNRs. Finally, we apply the compensation algorithm to a connected digit string recognition system that utilizes speech signals decoded by the GSM adaptive multi-rate (AMR) speech coder. It is shown that the noise-robust compensation algorithm reduces word error rates by 15% or more at low bit-rate modes of the AMR speech coder.
ER -