Proceedings Article10.21437/INTERSPEECH.2007-5
Soft margin feature extraction for automatic speech recognition.
Jinyu Li,Chin-Hui Lee +1 more
- 27 Aug 2007
- pp 30-33
TL;DR: This first study on applying the margin-based method in joint optimization of feature extraction and acoustic modeling demonstrates the success of soft margin based method, which targets to obtain both high accuracy and good model generalization.
read more
Abstract: We propose a new discriminative learning framework, called soft margin feature extraction (SMFE), for jointly optimizing the parameters of transformation matrix for feature extraction and of hidden Markov models (HMMs) for acoustic modeling. SMFE extends our previous work of soft margin estimation (SME) to feature extraction. Tested on the TIDIGITS connected digit recognition task, the proposed approach achieves a string accuracy of 99.61%, much better than our previously reported SME results. To our knowledge, this is the first study on applying the margin-based method in joint optimization of feature extraction and acoustic modeling. The excellent performance of SMFE demonstrates the success of soft margin based method, which targets to obtain both high accuracy and good model generalization.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
•Proceedings Article
Soft Margin Estimation of Hidden Markov Model Parameters (best student paper)
Jinyu Li
- 01 Jan 2006
TL;DR: In this article, the authors proposed a new discriminative learning framework, called soft margin estimation (SME), for estimating parameters of continuous density hidden Markov models, which makes direct use of the successful ideas of soft margin in support vector machines to improve generalization capability, and of decision feedback learning in minimum classification error training to enhance model separation in classifier design.
42
Training data selection for improving discriminative training of acoustic models
TL;DR: Experiments conducted on the Mandarin broadcast news collected in Taiwan shown that both phone-and frame-level data selection could achieve slight but consistent improvements over the baseline systems at lower training iterations.
22
Robust speech recognition under noisy ambient conditions
Kuldip K. Paliwal,Kaisheng Yao +1 more
- 01 Jan 2010
TL;DR: A brief overview of an automatic speech recognition system is provided, sources of speech variability that cause mismatch between training and testing are described, and some of the current techniques to achieve robust speech recognition are discussed.
•Dissertation
Soft margin estimation for automatic speech recognition
Chin-Hui Lee,Jinyu Li +1 more
- 01 Jan 2008
TL;DR: This is the first attempt to show the effectiveness of margin-based acoustic modeling for large vocabulary continuous speech recognition in a HMMs framework.
20
A study on soft margin estimation for LVCSR
Jinyu Li,Zhi-Jie Yan,Chin-Hui Lee,Ren-Hua Wang +3 more
- 01 Jan 2007
TL;DR: The extended Baum-Welch method is used to replace the conventional generalized probabilistic descent algorithm for optimization in soft margin estimation to large vocabulary continuous speech recognition in two aspects.
References
•Book
The Nature of Statistical Learning Theory
Vladimir Vapnik
- 01 Jan 1995
TL;DR: Setting of the learning problem consistency of learning processes bounds on the rate of convergence ofLearning processes controlling the generalization ability of learning process constructing learning algorithms what is important in learning theory?
46K
Comparison of parametric representations for monosyllabic word recognition in continuously spoken sentences
S. Davis,Paul Mermelstein +1 more
TL;DR: In this article, several parametric representations of the acoustic signal were compared with regard to word recognition performance in a syllable-oriented continuous speech recognition system, and the emphasis was on the ability to retain phonetically significant acoustic information in the face of syntactic and duration variations.
5.3K
Maximum mutual information estimation of hidden Markov model parameters for speech recognition
Lalit R. Bahl,Peter Fitzhugh Brown,P.V. de Souza,Robert Leroy Mercer +3 more
- 07 Apr 1986
TL;DR: A method for estimating the parameters of hidden Markov models of speech is described and recognition results are presented comparing this method with maximum likelihood estimation.
983
Minimum Phone Error and I-smoothing for improved discriminative training
Daniel Povey,Philip C. Woodland +1 more
- 13 May 2002
TL;DR: The Minimum Phone Error (MPE) and Minimum Word Error (MWE) criteria are smoothed approximations to the phone or word error rate respectively and I-smoothing which is a novel technique for smoothing discriminative training criteria using statistics for maximum likelihood estimation (MLE).
Testing Statistical Hypothesis.
C. F. Crouse,E. L. Lehmann +1 more
TL;DR: In this paper, the authors propose to test statistical hypothese in the context of testing statistical hypotheses, and test statistical hypotheses in the setting of statistical hypothesis testing, testing statistical hypothetse.
778
Related Papers (5)
Daniel Povey,Philip C. Woodland +1 more
- 13 May 2002
Vladimir Vapnik
- 01 Jan 1995
A. Biem,Shigeru Katagiri,Biing-Hwang Juang +2 more
- 06 Sep 1993