On compensating the Mel-frequency cepstral coefficients for noisy speech recognition

Authors:
Eric H. C. Choi
Affiliations:
Interfaces, Machines and Graphic Environments (IMAGEN), National ICT Australia, Alexandria, NSW, Sydney, Australia
Venue:
ACSC '06 Proceedings of the 29th Australasian Computer Science Conference - Volume 48
Year:
2006

Citing 4
Cited 1

Fundamentals of speech recognition

Fundamentals of speech recognition
The image processing handbook (2nd ed.)

The image processing handbook (2nd ed.)
Challenges in adopting speech recognition

Communications of the ACM - Multimodal interfaces that flex, adapt, and persist
Advanced Digital Signal Processing and Noise Reduction

Advanced Digital Signal Processing and Noise Reduction

Incorporating verbal feedback into a robot-assisted rehabilitation system

Robotica

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes a novel noise-robust automatic speech recognition (ASR) front-end that employs a combination of Mel-filterbank output compensation and cumulative distribution mapping of cepstral coefficients with truncated Gaussian distribution. Recognition experiments on the Aurora II connected digits database reveal that the proposed front-end achieves an average digit recognition accuracy of 84.92% for a model set trained from clean speech data. Compared with the ETSI standard Mel-cepstral front-end, the proposed front-end is found to obtain a relative error rate reduction of around 61%. Moreover, the proposed front-end can provide comparable recognition accuracy with the ETSI advanced front-end, at less than half the computation load.