HMM2- Extraction of Formant Features and their Use for Robust ASR

As recently introduced, an HMM2 can be considered as a particular case of an HMM mixture in which the HMM emission probabilities (usually estimated through Gaussian mixtures or an artificial neural network) are modeled by state-dependent, feature-based HMM (referred to as frequency HMM). A general EM training algorithm for such a structure has already been developed. Although there are numerous motivations for using such a structure, and many possible ways to exploit it, this paper will mainly focus on one particular instantiation of HMM2 in which the frequency HMM will be used to extract formant structure information, which will then be used as additional acoustic features in a standard Automatic Speech Recognition (ASR) system. While the fact that this architecture is able to automatically extract meaningful formant information is interesting by itself, empirical results will also show the robustness of these features to noise, and their potential to enhance state-of-the-art noise-robust HMM-based ASR.

[1]  Samy Bengio,et al.  HMM2- a novel approach to HMM emission probability estimation , 2000, INTERSPEECH.

[2]  Philip N. Garner,et al.  On the robust incorporation of formant features into hidden Markov models for automatic speech recognition , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[3]  Gerhard Rigoll,et al.  High performance face recognition using pseudo 2-D hidden Markov models , 1999, 1999 European Control Conference (ECC).

[4]  Ronald A. Cole,et al.  New telephone speech corpora at CSLU , 1995, EUROSPEECH.

[5]  Ferdinando Silvestro Samaria,et al.  Face recognition using Hidden Markov Models , 1995 .

[6]  Hermann Ney,et al.  Formant estimation for speech recognition , 1998, IEEE Trans. Speech Audio Process..

[7]  Samy Bengio,et al.  An EM Algorithm for HMMs with Emission Distributions Represented by HMMs , 2000 .

[8]  Gary E. Kopec.  Formant tracking using hidden Markov models and vector quantization , 1986, IEEE Trans. Acoust. Speech Signal Process..

[9]  Katrin Weber.  Multiple Timescale Feature Combination Towards Robust Speech Recognition , 2000, KONVENS.

[10]  Steve Young,et al.  The HTK book , 1995 .

[11]  Oscar E. Agazzi,et al.  Machine vision for keyword spotting using pseudo 2D hidden Markov models , 1993, 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing.