CONTINUOUS SPEECH RECOGNITION USING SYLLABIC SEGMENTATION AND DEMISYLLABLE HIDDEN MARKOV MODELS

1st European Conference on Speech Communication and Technology, EUROSPEECH 1989

Volumn , Issue , 1989, Pages 1017-1020

  Weigel, Walter ^a   Ruske, Günther ^a  

^a TECHNICAL UNIVERSITY OF MUNICH   (Germany)

Author keywords

[No Author keywords available]

Indexed keywords

SPEECH COMMUNICATION; SPEECH RECOGNITION; TRELLIS CODES;

CO-ARTICULATION; DECODING MODULE; DISCRETE HIDDEN MARKOV MODEL; GAUSSIAN-HMM; HIDDEN-MARKOV MODELS; LOCALISATION; RECOGNITION UNITS; SPEECH RECOGNITION SYSTEMS; SYLLABLE BOUNDARIES; VECTOR QUANTISATION;

HIDDEN MARKOV MODELS;

EID: 84892165149     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (3)

1. ZWICKER, E., TERHARDT, E., and PAULUS, E., Automatie speech recognition using psychoacoustic models, JASA 65 (2), 1979, 487-489.
2. RUSKE, G. und WEIGEL, W., Automatie recognition of spoken sentences using a demisyllablebased Dynamic Programming algori thm. 12th International Congress on Acoustics, Toronto 1986, paper Al-5.
3. RUSKE, G., Experiments on the use of demisyllables for automatic speech recognition. Symp. on Speech Recognition, Montreal 1986, McGill Universi ty, 49-50.