Hybrid NN/HMM acoustic modeling techniques for distributed speech recognition

Speech Communication

Volumn 48, Issue 8, 2006, Pages 1037-1046

  Stadermann, Jan ^a   Rigoll, Gerhard ^a  

^a TECHNICAL UNIVERSITY OF MUNICH   (Germany)

Author keywords

Distributed speech recognition; Hybrid speech recognition; Tied posteriors

Indexed keywords

ACOUSTIC EQUIPMENT; MATHEMATICAL MODELS; SPEECH ANALYSIS; SPEECH SYNTHESIS;

DISTRIBUTED SPEECH RECOGNITION (DSR); GAUSSIAN MODELS; HYBRID ACOUSTIC MODELS; HYBRID SPEECH RECOGNITION; RASTA-PLP FEATURES; TIED-POSTERIORS;

SPEECH RECOGNITION;

EID: 33745351889     PISSN: 01676393     EISSN: None     Source Type: Journal    
DOI: 10.1016/j.specom.2006.01.007     Document Type: Article

References (17)

1. Barras, C., Lamel, L., Gauvain, J., 2001. Automatic transcription of compressed broadcast audio. In: IEEE Int. Conf. on Acoustics, Speech, and Signal Processing (ICASSP), Salt Lake City, UT, USA.
2. Bourlard H., and Morgan N. Connectionist Speech Recognition: A Hybrid Approach (1994), Kluwer Academic Publishers
3. Bourlard H., and Wellekens C. Links between Markov models and multilayer perceptrons. IEEE Trans. Pattern Anal. Machine Intell. 12 December (1990) 1167-1178
4. Hermansky H., and Morgan N. RASTA processing of speech. IEEE Trans. Speech Audio Process. 2 4 (1994) 578-589
5. Hirsch, H.G., Pearce, D., 2000. The AURORA experimental framework for the performance evaluation of speech recognition systems under noisy conditions. In: ISCA ITRW ASR2000.
6. James, A., Gómez, A., Milner, B., 2005. A comparison of packet loss compensation methods and interleaving for speech recognition in burst-like packet loss. In: IEEE Int. Conf. on Acoustics, Speech, and Signal Processing (ICASSP), Philadelphia, PA, USA, March.
7. Paul, D.B., Baker, J.M., 1992. The design for the wall street journal-based CSR corpus. In: Int. Conf. on Spoken Language Processing (ICSLP), Banff, Canada, October, pp. 899-902.
8. Rottland, J., Rigoll, G., 2000. Tied posteriors: an approach for effective introduction of context dependency in hybrid NN/HMM LVCSR. In: IEEE Int. Conf. on Acoustics, Speech, and Signal Processing (ICASSP), Istanbul, Turkey, June.
9. Santini S., and Del Bimbo A. Recurrent neural networks can be trained to be maximum a posteriori probability classifiers. Neural Networks 8 1 (1995) 25-29
10. Schulz, H., Mast, M., Ross, T., Harrikari, H., Stadermann, J., Demesticha, V., Vamvakoulas, Y., Polymenakos, L., 2001. A conversational natural language understanding information system for multiple languages. In: 6th Int. Workshop on Applications of Natural Language for Information Systems, Madrid, Spain, June.
11. Stadermann, J., Rigoll, G., 2001. Comparison of standard and hybrid modeling techniques for distributed speech recognition. In: IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU), Madonna di Campiglio Trento, Italy, December.
12. Stadermann, J., Rigoll, G., 2003a. Comparing NN paradigms in hybrid NN/HMM speech recognition using tied posteriors. In: IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU), November, St. Thomas, U.S. Virgin Islands, November.
13. Stadermann, J., Rigoll, G., 2003b. Flexible feature extraction and HMM design for a hybrid distributed speech recognition system in noisy environments. In: IEEE Int. Conf. on Acoustics, Speech, and Signal Processing (ICASSP), Hong Kong, China, April.
14. Stadermann, J., Rigoll, G., 2005. Verteilte Spracherkennung mit hybriden akustischen Modellen. In: 3.1. Deutsche Jahrestagung fnr Akustik, DAGA05, Mnnchen, Deutschland, March.
15. Stadermann, J., Meermeier, R., Rigoll, G., 2001. Distributed speech recognition using traditional and hybrid modeling techniques. In: European Conf. on Speech Communication and Technology, Aalborg, Denmark, September.
16. Speech processing, transmission and quality aspects (STQ), 2003. Distributed speech recognition; front-end feature extraction algorithm; compression algorithms. In: ETSI ES 201 108 v1.1.3 (2003-09), September.
17. Yuk, D., Flanagan, J., 1999. Telephone speech recognition using neural networks and hidden markov models. In: IEEE Int. Conf. on Acoustics, Speech, and Signal Processing (ICASSP), 1999.