Canonicalization of Feature Parameters for Robust Speech Recognition Based on Distinctive Phonetic Feature (DPF) Vectors
Summary :
This paper describes a robust automatic speech recognition (ASR) system with less computation. Acoustic models of a hidden Markov model (HMM)-based classifier include various types of hidden factors such as speaker-specific characteristics, coarticulation, and an acoustic environment, etc. If there exists a canonicalization process that can recover the degraded margin of acoustic likelihoods between correct phonemes and other ones caused by hidden factors, the robustness of ASR systems can be improved. In this paper, we introduce a canonicalization method that is composed of multiple distinctive phonetic feature (DPF) extractors corresponding to each hidden factor canonicalization, and a DPF selector which selects an optimum DPF vector as an input of the HMM-based classifier. The proposed method resolves gender factors and speaker variability, and eliminates noise factors by applying the canonicalzation based on the DPF extractors and two-stage Wiener filtering. In the experiment on AURORA-2J, the proposed method provides higher word accuracy under clean training and significant improvement of word accuracy in low signal-to-noise ratio (SNR) under multi-condition training compared to a standard ASR system with mel frequency ceptral coeffient (MFCC) parameters. Moreover, the proposed method requires a reduced, two-fifth, Gaussian mixture components and less memory to achieve accurate ASR.
- Publication
- IEICE TRANSACTIONS on Information Vol.E91-D No.3 pp.488-498
- Publication Date
- 2008/03/01
- Publicized
- Online ISSN
- 1745-1361
- DOI
- 10.1093/ietisy/e91-d.3.488
- Type of Manuscript
- Special Section PAPER (Special Section on Robust Speech Processing in Realistic Environments)
- Category
- Feature Extraction
Authors
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Mohammad NURUL HUDA, Muhammad GHULAM, Takashi FUKUDA, Kouichi KATSURADA, Tsuneo NITTA, "Canonicalization of Feature Parameters for Robust Speech Recognition Based on Distinctive Phonetic Feature (DPF) Vectors" in IEICE TRANSACTIONS on Information,
vol. E91-D, no. 3, pp. 488-498, March 2008, doi: 10.1093/ietisy/e91-d.3.488.
Abstract: This paper describes a robust automatic speech recognition (ASR) system with less computation. Acoustic models of a hidden Markov model (HMM)-based classifier include various types of hidden factors such as speaker-specific characteristics, coarticulation, and an acoustic environment, etc. If there exists a canonicalization process that can recover the degraded margin of acoustic likelihoods between correct phonemes and other ones caused by hidden factors, the robustness of ASR systems can be improved. In this paper, we introduce a canonicalization method that is composed of multiple distinctive phonetic feature (DPF) extractors corresponding to each hidden factor canonicalization, and a DPF selector which selects an optimum DPF vector as an input of the HMM-based classifier. The proposed method resolves gender factors and speaker variability, and eliminates noise factors by applying the canonicalzation based on the DPF extractors and two-stage Wiener filtering. In the experiment on AURORA-2J, the proposed method provides higher word accuracy under clean training and significant improvement of word accuracy in low signal-to-noise ratio (SNR) under multi-condition training compared to a standard ASR system with mel frequency ceptral coeffient (MFCC) parameters. Moreover, the proposed method requires a reduced, two-fifth, Gaussian mixture components and less memory to achieve accurate ASR.
URL: https://global.ieice.org/en_transactions/information/10.1093/ietisy/e91-d.3.488/_p
Copy
@ARTICLE{e91-d_3_488,
author={Mohammad NURUL HUDA, Muhammad GHULAM, Takashi FUKUDA, Kouichi KATSURADA, Tsuneo NITTA, },
journal={IEICE TRANSACTIONS on Information},
title={Canonicalization of Feature Parameters for Robust Speech Recognition Based on Distinctive Phonetic Feature (DPF) Vectors},
year={2008},
volume={E91-D},
number={3},
pages={488-498},
abstract={This paper describes a robust automatic speech recognition (ASR) system with less computation. Acoustic models of a hidden Markov model (HMM)-based classifier include various types of hidden factors such as speaker-specific characteristics, coarticulation, and an acoustic environment, etc. If there exists a canonicalization process that can recover the degraded margin of acoustic likelihoods between correct phonemes and other ones caused by hidden factors, the robustness of ASR systems can be improved. In this paper, we introduce a canonicalization method that is composed of multiple distinctive phonetic feature (DPF) extractors corresponding to each hidden factor canonicalization, and a DPF selector which selects an optimum DPF vector as an input of the HMM-based classifier. The proposed method resolves gender factors and speaker variability, and eliminates noise factors by applying the canonicalzation based on the DPF extractors and two-stage Wiener filtering. In the experiment on AURORA-2J, the proposed method provides higher word accuracy under clean training and significant improvement of word accuracy in low signal-to-noise ratio (SNR) under multi-condition training compared to a standard ASR system with mel frequency ceptral coeffient (MFCC) parameters. Moreover, the proposed method requires a reduced, two-fifth, Gaussian mixture components and less memory to achieve accurate ASR.},
keywords={},
doi={10.1093/ietisy/e91-d.3.488},
ISSN={1745-1361},
month={March},}
Copy
TY - JOUR
TI - Canonicalization of Feature Parameters for Robust Speech Recognition Based on Distinctive Phonetic Feature (DPF) Vectors
T2 - IEICE TRANSACTIONS on Information
SP - 488
EP - 498
AU - Mohammad NURUL HUDA
AU - Muhammad GHULAM
AU - Takashi FUKUDA
AU - Kouichi KATSURADA
AU - Tsuneo NITTA
PY - 2008
DO - 10.1093/ietisy/e91-d.3.488
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E91-D
IS - 3
JA - IEICE TRANSACTIONS on Information
Y1 - March 2008
AB - This paper describes a robust automatic speech recognition (ASR) system with less computation. Acoustic models of a hidden Markov model (HMM)-based classifier include various types of hidden factors such as speaker-specific characteristics, coarticulation, and an acoustic environment, etc. If there exists a canonicalization process that can recover the degraded margin of acoustic likelihoods between correct phonemes and other ones caused by hidden factors, the robustness of ASR systems can be improved. In this paper, we introduce a canonicalization method that is composed of multiple distinctive phonetic feature (DPF) extractors corresponding to each hidden factor canonicalization, and a DPF selector which selects an optimum DPF vector as an input of the HMM-based classifier. The proposed method resolves gender factors and speaker variability, and eliminates noise factors by applying the canonicalzation based on the DPF extractors and two-stage Wiener filtering. In the experiment on AURORA-2J, the proposed method provides higher word accuracy under clean training and significant improvement of word accuracy in low signal-to-noise ratio (SNR) under multi-condition training compared to a standard ASR system with mel frequency ceptral coeffient (MFCC) parameters. Moreover, the proposed method requires a reduced, two-fifth, Gaussian mixture components and less memory to achieve accurate ASR.
ER -
English
