5th International Conference on Spoken Language Processing

Sydney, Australia
November 30 - December 4, 1998

Improved Parallel Model Combination Based on Better Domain Transformation for Speech Recognition Under Noisy Environments

Jeih-Weih Hung (1), Jia-Lin Shen (1), Lin-Shan Lee (2)

(1) IIS, Academia Sinica, Taiwan
(2) the department of E.E., National Taiwan University, Taiwan

The parallel model combination (PMC) technique has been shown to achieve very good performance for speech recognition under noisy conditions. However, there still exist some problems based on the PMC formula. In this paper, we first investigated these problems and some modifications on the transformation process of PMC were proposed. Experimental results show that this modified PMC can provide significant improvements over the original PMC in the recognition accuracies. Error rate reduction on the order of 12.92% was achieved.

Full Paper

Bibliographic reference.  Hung, Jeih-Weih / Shen, Jia-Lin / Lee, Lin-Shan (1998): "Improved parallel model combination based on better domain transformation for speech recognition under noisy environments", In ICSLP-1998, paper 0231.