Robust excitation-based features for Automatic Speech Recognition
Authors:
- 1 April 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15206149,p. 4664-4668
Abstract
In this paper we investigate the use of noise-robust features characterizing the speech excitation signal as complementary features to the usually considered vocal tract based features for Automatic Speech Recognition (ASR). The proposed Excitation-based Features (EBF) are tested in a state-of-the-art Deep Neural Network (DNN) based hybrid acoustic model for speech recognition. The suggested excitation features expand the set of periodicity features previously considered for ASR, expecting that these features help in a better discrimination of the broad phonetic classes (e.g., fricatives, nasal, vowels, etc.). Our experiments on the AMI meeting transcription system showed that the proposed EBF yield a relative word error rate reduction of about 5% when combined with conventional PLP features. Further experiments led on Aurora4 confirmed the robustness of the EBF to both additive and convolutive noises, with a relative improvement of 4.3% obtained by combinining them with mel filter banks.Keywords
This publication has 27 references indexed in Scilit:
- Glottal source processing: From analysis to applicationsComputer Speech & Language, 2014
- A comparative study of glottal source estimation techniquesComputer Speech & Language, 2012
- Chirp group delay analysis of speech signalsSpeech Communication, 2007
- The AMI Meeting Corpus: A Pre-announcement2006
- Use of periodicity and jitter as speech recognition features2002
- Use of voicing features in HMM-based speech recognitionSpeech Communication, 2002
- Robust voice activity detection using higher-order statistics in the LPC residual domainIEEE Transactions on Speech and Audio Processing, 2001
- Acoustic Correlates of Breathy Vocal Quality: Dysphonic Voices and Continuous SpeechJournal of Speech, Language, and Hearing Research, 1996
- Perceptual linear predictive (PLP) analysis of speechThe Journal of the Acoustical Society of America, 1990
- Comparison of parametric representations for monosyllabic word recognition in continuously spoken sentencesIEEE Transactions on Acoustics, Speech, and Signal Processing, 1980