Improving bottleneck features for Vietnamese large vocabulary continuous speech recognition system using deep neural networks

Bao Quoc Nguyen, Thang Tat Vu, Mai Chi Luong
Author affiliations

Authors

  • Bao Quoc Nguyen
  • Thang Tat Vu Institute of Information Technology, Vietnam Academy of Science and Technology
  • Mai Chi Luong Institute of Information Technology, Vietnam Academy of Science and Technology

DOI:

https://doi.org/10.15625/1813-9663/31/4/5944

Keywords:

Deep bottleneck features, neural network, Vietnamese speech recognition.

Abstract

In this paper, the pre-training method based on denoising auto-encoder is investigated and proved to be good models for initializing bottleneck networks of Vietnamese speech recognition system that result in better recognition performance compared to base bottleneck features reported previously. The experiments are carried out on the dataset containing speeches on Voice of Vietnam channel (VOV). The results show that the DBNF extraction for Vietnamese recognition decreases relative word error rate by 14 % and 39 % compared to the base bottleneck features and MFCC baseline, respectively.

Metrics

Metrics Loading ...

Downloads

Published

03-01-2016

How to Cite

[1]
B. Q. Nguyen, T. T. Vu, and M. C. Luong, “Improving bottleneck features for Vietnamese large vocabulary continuous speech recognition system using deep neural networks”, JCC, vol. 31, no. 4, p. 267, Jan. 2016.

Issue

Section

Computer Science

Most read articles by the same author(s)