Robust Speech Recognition Using A Voiced-Unvoiced Feature (2002)
| Venue: | IN PROC. INT. CONF. ON SPOKEN LANGUAGE PROCESSING |
| Citations: | 5 - 2 self |
BibTeX
@INPROCEEDINGS{Zolnay02robustspeech,
author = {András Zolnay and Ralf Schlüter and Hermann Ney},
title = {Robust Speech Recognition Using A Voiced-Unvoiced Feature},
booktitle = {IN PROC. INT. CONF. ON SPOKEN LANGUAGE PROCESSING},
year = {2002},
pages = {1065--1068},
publisher = {}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
In this paper, a voiced-unvoiced measure is used as acoustic feature for continuous speech recognition. The voiced-unvoiced measure was combined with the standard Mel Frequency Cepstral Coefficients (MFCC) using linear discriminant analysis (LDA) to choose the most relevant features. Experiments were performed on the SieTill (German digit strings recorded over telephone line) and on the SPINE (English spontaneous speech under different simulated noisy environments) corpus. The additional voiced-unvoiced measure results in improvements in word error rate (WER) of up to 11% relative to using MFCC alone with the same overall number of parameters in the system.
Citations
| 14 | Perceptual harmonic cepstral coefficients for speech recognition in noisy environments - Gu, Rose - 2001 |
| 11 | Cepstrum-Based Pitch Detection Using a New Statisctical V/UV Classification Algorithm - Ahmadi, Spanias - 1999 |
| 11 | Incorporating voice onset time to improve letter recognition accuracies - Niyogi, Ramesh - 1998 |
| 8 | Comparison of Optimization Methods for Discriminative Training Criteria - Schlüter, Macherey, et al. - 1997 |
| 5 | Rabiner and B-H - R - 1993 |
| 3 | Robust Speech Recognition Techniques Applied to a Speech - Rose, Kim, et al. - 2001 |
| 1 | Multistage Coarticulatory Model Combining Articulatory, Formant and Cepstral Features - Gao, Bakis, et al. - 2000 |
