Evaluation of the stochastic morphosyntactic language model on a one millionword Hungarian dictation task

EUROSPEECH 2003 - 8th European Conference on Speech Communication and Technology

Volumn , Issue , 2003, Pages 2297-2300

  Szarvas, Máté ^a   Furui, Sadaoki ^a  

^a TOKYO INSTITUTE OF TECHNOLOGY   (Japan)

Author keywords

[No Author keywords available]

Indexed keywords

ERRORS; NETWORK ARCHITECTURE; SPEECH COMMUNICATION; SPEECH RECOGNITION; STOCHASTIC MODELS; STOCHASTIC SYSTEMS; TRANSDUCERS;

COMBINED MODEL; HUNGARIANS; LANGUAGE MODEL; N-GRAM MODELING; RECOGNITION SYSTEMS; RECOGNITION UNITS; WEIGHTED FINITE-STATE TRANSDUCERS; WORD ERROR RATE;

COMPUTATIONAL LINGUISTICS;

EID: 65349171767     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (4)

1. L. Gillick and S. Cox. Some Statistical Issues in the Comparison of SpeechRecognition Algorithms. In ICASSP 89, pp. 532-535.
2. M. Mohri, F. Pereira, M. Riley. Weighted Finite-State Transducers in Speech Recognition. In Proc. ISCA Automatic Speech Recognition 2000., pp. 97-106.
3. M. Szarvas, S. Furui. Finite-state transducer based Hungarian LVCSR with explicit modeling of phonological changes. In Proc. ICSLP 2002., pp. 1297-1300.
4. M. Szarvas, S. Furui. Finite-state transducer based modeling of morphosyntax with applications to Hungarian LVCSR. To appear in Proc. ICASSP 2003.