Advanced Topics for Class Presentation
(Once
you form a group, choose a topic as soon as you can. I assign topics on a
first-ask-first-get basis.) (Marked * means the topic is taken by others.
1-person group is allowed only for last person if total number is odd.)
*
[1] MAP estimation of HMM: J.-L. Gauvain
and C.-H. Lee, ¡°Maximum a posteriori
Estimation for Multivariate Gaussian mixture Observation of Markov Chains,¡±
IEEE Trans. on Speech and Audio Processing, pp.291-298, Vol.2, 1994.
[2]
On-Line Bayesian Learning of HMM: Q.Huo and
C.-H. Lee, ¡°On-line Adaptive Learning of the
Continuous Density Hidden Markov Model based on Approximate Recursive Bayes
Estimate,¡± IEEE Trans. on Speech and Audio Processing, pp.161-172, Vol. 5,
No. 2, 1997.
[3]
MMIE of HMM: Y. Normandin, R. Cardin and R. De Mori, ¡°High-Performance Connected Digit Recognition
Using Maximum Mutual Information Estimation,¡± IEEE Trans. on Speech and
Audio Processing, pp. Vol. 2, No. 2, Apr. 1994.
[4]
MCE of HMM: B.-H.
Juang, W. Chou and C.-H. Lee, ¡°Minimum
Classification Error Rate Methods for Speech Recognition,¡± IEEE Trans. on
Speech and Audio Processing, pp.257-265, Vol. 5, No. 3, May 1997.
[5]
Large Marge Estimation of HMM: H. Jiang, X. Li and C.-J. Liu, ¡°Large Margin Hidden Markov Models for Speech
Recognition,¡± IEEE Trans. On Audio, Speech and Language Processing,?
pp.1584-1595, Vol. 14, No. 5, September 2006.
[6]
Decision Tree for HMM tying:
[7]
Search in large vocabulary ASR:
[8]
Adaptive Statistical Language Modeling: R. Rosenfeld, ¡°A maximum entropy approach to adaptive
statistical language modeling,¡± Computer Speech and Language, Vol. 10,
pp.187-228, 1996.
[9]
Transformation-based speaker adaptation: C. J.
Leggetter and P. C. Woodland, ¡°Maximum
Likelihood Linear Regression for Speaker Adaptation of Continuous Density
Hidden Markov Models,¡± Computer Speech and Language, pp. 171-185, Vol. 9,
1995.
*[10]
Speaker Verification: Q. Li, B.-H. Juang, C.-H. Lee, ¡°Automatic verbal information verification for user
authentication¡± IEEE Trans. on
Speech and Audio Processing, pp.585-596, Vol. 8, No. 5, Sep. 2000.
[11]
Utterance Verification (outlier rejection): M. G. Rahim,
C.-H. Lee and B.-H. Juang, ¡°Discriminative
Utterance Verification for Connected Digits Recognition,¡± IEEE Trans. on
Speech and Audio Processing,
pp.266-277, Vol. 5, No.3,
May 1997.
[12]
Bayesian Approach to Speaker Verification: H. Jiang and L. Deng, ¡°A Bayesian Approach to the
Verification problem: Applications
to Speaker Verification,¡± IEEE Trans. on Speech and Audio Processing, pp.874-884, Vol. 9, No. 8, Nov.
2001.
[13]
Latent Semantic Analysis for LM: J. R. Bellegarda, ¡°Exploiting Latent Semantic Information in
Statistical Language
Modeling,¡± Proceedings of IEEE, pp.1279-1296, Vol.88, No. 8, August
2000.
*[14]
Speech Understanding: T. Kawahara, C.-H. Lee and B.-H. Juang, ¡°Flexible speech understanding based
on combined key-phrase detection and verification,¡± IEEE Trans. on Speech
and Audio Processing, pp.558-568, Vol. 6, No. 6, November 1998.
[15]
Statistical machine translation: P. F. Brown, et. al., ¡°The Mathematics of Statistical Machine
Translation: Parameter
Estimation,¡± Computational Linguistics, pp.263-297, Vol. 19, No. 2, 1993.
[16]
Support Vector Machines: C. J. Burges, ¡°A Tutorial on Support Vector Machines for
Pattern Recognition,¡± Data Mining and Knowledge Discovery, 2, 121-167 (1998).
*[17]
Weighted Finite State Transducer (WFST) Optimization: M. Mohri, F.
Pereira and M. Riley, ¡°Speech Recognition with
Weighted Finite-State Transducers,¡± In Larry Rabiner and Fred Juang, editors,
Handbook on Speech Processing and Speech Communication, Part E: Speech
recognition. Springer-Verlag,
*[18]
Graphical models: Graphical models, Chapter 8
of Pattern Recognition and Machine Learning by Christopher M. Bishop.
[19]
Boosting: Y. Freund and R. E. Schapire, ¡°A Decision-Theoretic
Generalization of On-Line Learning and an Application to Boosting,¡± Journal
of computer and system sciences 55, 119-139 (1997).