Fast LM look-ahead for large vocabulary continuous speech recognition using perfect hashing

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

Volumn 1, Issue , 2002, Pages 705-708

  Cardenal López, Antonio ^a   Diéguez Tirado, F Javier ^a   García Mateo, Carmen ^a  

^a UNIVERSITY OF VIGO   (Spain)

Author keywords

[No Author keywords available]

Indexed keywords

ALGORITHMS; APPROXIMATION THEORY; CACHE MEMORY; CALCULATIONS; DATABASE SYSTEMS; MARKOV PROCESSES; MATHEMATICAL MODELS; NATURAL LANGUAGE PROCESSING SYSTEMS; SPEECH CODING; TREES (MATHEMATICS);

DECODER; HASHING; HIDDEN MARKO MODEL; LANGUAGE MODEL LOOKAHEAD ALGORITHM; VITERBI SYNCHRONUOUS BEAM SEARCH;

CONTINUOUS SPEECH RECOGNITION;

EID: 4544242764     PISSN: 15206149     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/icassp.2002.5743815     Document Type: Conference Paper

References (7)

1. A one pass decoder design for large vocabulary recognition
2. The 1999 CMU 10x real time broadcast news transcription system
3. Improved lexical tree search for large vocabulary speech recognition
4. An efficient search space representation for large vocabulary continuous speech recognition
5. The demiphone: An efficient subword unit for continuous speech recognition
6. Storing a sparse table with O(1) worst case access time
7. An optimal algorithm for generating minimal perfect hash functions