MINIMUM CONFUSIBILITY TRAINING OF CONTEXT DEPENDENT DEMIPHONES

6th European Conference on Speech Communication and Technology, EUROSPEECH 1999

Volumn , Issue , 1999, Pages 2741-2744

  Nogueiras Rodríguez, Albino ^a   Mariño, Jose B ^a  

^a UNIVERSITAT POLITÈCNICA DE CATALUNYA   (Spain)

Author keywords

[No Author keywords available]

Indexed keywords

ACOUSTIC MODELING; SPEECH COMMUNICATION; SPEECH RECOGNITION; TELEPHONE SETS;

% REDUCTIONS; ACOUSTICS MODEL; CONTEXT DEPENDENT; DISCRIMINATIVE TRAINING; ERROR RATE; MA XIMUM LIKELIHOODS; MAXIMUM-LIKELIHOOD; SPEECH RECOGNITION SYSTEMS; SUBWORD UNITS; TRAINING ALGORITHMS;

MAXIMUM LIKELIHOOD;

EID: 85017302607     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (5)

1. C.-H. Lee, B.-H. Juang, W. Chou, and J.J. Molina. A study on task-independent subword selection and modeling for speech recognition. In Proc. of ICLSP'96, pages 1820-1823, 1996.
2. J.B. Marino, A. Nogueiras, and A. Bonafonte. The demiphone: an efficient subword unit for continuous speech recognition. In Proc. of EUROSPEECH'97, pages 1215-1218, 1997.
3. J.B. Marino, P. Paches-Leal, and A. Nogueiras. The demiphone versus the triphone in a decision-tree state-tying framework. In Proc. of ICSLP'98, 1998.
4. A. Nogueiras and J.B. Marino. Task adaptation of sub-lexical unit models using the minimum confusibility criterion on task independent databases. In Proc. of ICSLP'98, September 1998.
5. A. Nogueiras and J.B. Marino. Task independent minimum cofusibility training for continuous speech recognition. In Proc. of ICASSP'98, pages 477-480, May 1998.