Incorporating tone-related MLP posteriors in the feature representation for Mandarin ASR

9th European Conference on Speech Communication and Technology

Volumn , Issue , 2005, Pages 2981-2984

  Lei, Xin ^a   Hwang, Mei Yuh ^a   Ostendorf, Mari ^a  

^a UNIVERSITY OF WASHINGTON   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

AMBIGUOUS WORDS; AUTOMATIC SPEECH RECOGNITION SYSTEMS; MANDARIN SPEECH; MLP POSTERIORS;

ACOUSTICS; FORMAL LANGUAGES; MARKOV PROCESSES; MATHEMATICAL MODELS; MULTILAYER NEURAL NETWORKS; SPEECH RECOGNITION; WORD PROCESSING;

FEATURE EXTRACTION;

EID: 33745210540     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (10)

1. C.J. Chen et al., "New methods in continuous Mandarin speech recognition," in Proc. Eur. Conf. Speech Communication Technology, 1997, vol. 3, pp. 1543-1546.
2. E. Chang et al., "Large vocabulary Mandarin speech recognition with different approaches in modeling tones," in Proc. ICSLP, 2000, vol. 2, pp. 983-986.
3. H.C. Huang and F. Seide, "Pitch tracking and tone features for Mandarin speech recognition," in Proc. ICASSP, 2000, vol. 3, pp. 1523-1526.
4. M. Hwang et al., "Porting DECIPHER from English to Mandarin," in Proc. DARPA 2004 Rich Transcriptions Workshop, 2004.
5. M.J.F. Gales et al., "Development of the CUHTK 2004 Mandarin conversational telephone speech transcription system," in Proc. ICASSP, 2005, vol. 1, pp. 841-844.
6. C. Wang, Prosodic Modeling for Improved Speech Recognition and Understanding, Ph.D. thesis, Massachusetts Institute of Technology, 2001.
7. H. Hermansky, D. Ellis, and S. Sharma, "Tandem connectionist feature extraction for conventional HMM systems," in Proc. ICASSP, 2000, pp. 1635-1638.
8. N. Morgan, B. Chen, Q. Zhu, and A. Stolcke, "TRAPping conversational speech: Extending TRAP/Tandem approaches to conversational telephone speech recognition," in Proc. ICASSP, 2004, vol. 1, pp. 537-540.
9. B. Chen, Q. Zhu, and N. Morgan, "Tonotopic multilayered perceptron: a neural network for learning longterm temporal features for speech recognition," in Proc. ICASSP, 2005, vol. 1, pp. 945-948.
10. M.K. Sonmez, L. Heck, M. Weintraub, and E. Shriberg, "A lognormal model of pitch for prosody-based speaker recognition," in Proc. Eur. Conf. Speech Communication Technology, 1997, vol. 3, pp. 1391-1394.