2000 IEEE International Conference on Acoustics, Speech, and Signal Processing: Silver Anniversary, Proceedings, 5-9 June 2000, Hilton Hotel and Convention Center, Istanbul, Turkey, Volume 3IEEE, 2000 - Electro-acoustics |
From inside the book
Results 1-3 of 47
Page 1568
... pronunciations were reflected into dictionary to distinguish Korean and English words with the same pronunciation . We also converted special symbols into their pronunciations . For example , ' - ' and ' between two numerals have been ...
... pronunciations were reflected into dictionary to distinguish Korean and English words with the same pronunciation . We also converted special symbols into their pronunciations . For example , ' - ' and ' between two numerals have been ...
Page 1679
... PRONUNCIATION AMBIGUITY VS PRONUNCIATION VARIABILITY IN SPEECH RECOGNITION Murat SaraƧlar and Sanjeev Khudanpur Center for Language and Speech Processing Johns Hopkins University , Baltimore , MD 21218-2686 { murat , sanjeev } @ clsp ...
... PRONUNCIATION AMBIGUITY VS PRONUNCIATION VARIABILITY IN SPEECH RECOGNITION Murat SaraƧlar and Sanjeev Khudanpur Center for Language and Speech Processing Johns Hopkins University , Baltimore , MD 21218-2686 { murat , sanjeev } @ clsp ...
Page 1777
... PRONUNCIATION MODELING Due to pronunciation reduction , some short phones in normal pronunciation can be deleted to fit fast speech better [ 1 ] . We have also observed the minimum duration problem of our HMM models in Fig.1 , which ...
... PRONUNCIATION MODELING Due to pronunciation reduction , some short phones in normal pronunciation can be deleted to fit fast speech better [ 1 ] . We have also observed the minimum duration problem of our HMM models in Fig.1 , which ...
Common terms and phrases
ACELP acoustic models adaptation algorithm amplitude analysis applied approach audio band baseline beam search bigram bits CELP cepstral clean speech clustering codebook coder components computed context corpus database decoder digit domain encoder estimation evaluation experiments extraction feature vectors Figure filter formant frame frequency function Gaussian graph harmonic hidden Markov models ICASSP IEEE improvement input interpolation iteration kbps language model lexical lexicon likelihood MELP method MFCC microphone mixture MLLR n-gram noise noisy obtained optimal output parameters perceptual performance phase phoneme pitch period probability Proc pronunciation proposed recognition accuracy reduced robust samples score sentences sequence Signal Processing sinusoidal speaker speaker recognition spectral spectrum speech coding speech recognition system speech segments speech signal syllable Table technique Technology tion training data transcription transform trigram triphone unigram University unvoiced utterances vector quantization vocabulary voiced waveform wavelet word error rate