What people are saying - Write a review
We haven't found any reviews in the usual places.
Classiﬁed Nonlinear Predictive Vector Quantization of Speech Spectral Parameters II761
Discriminative Training of Gaussian Mixture Models for Large Vocabulary Speech Recognition Systems II613
65 other sections not shown
Other editions - View all
accuracy acoustic models adaptation algorithm analysis bigram cepstral cepstrum classiﬁcation cluster codebook coder coefficients computed context continuous speech corpus covariance matrix database decoding deﬁned digit dynamic estimate evaluation experiments feature Figure ﬁlter ﬁnal ﬁrst ﬁxed frame frequency function Gaussian Hidden Markov Models ICASSP identiﬁcation IEEE IEEE Trans improvement input interpolation language model likelihood linear matrix mean vector measure MELP method microphone mixture n-gram neural network node noise noisy obtained optimal output parameters phoneme pitch probability Proc proposed quantization recognition rate recognizer reduced robust samples score segment semantic sentences sequence Signal Processing signiﬁcant speaker speaker recognition speaker-independent speciﬁc spectral spectrum speech coding speech data speech recognition system speech signal speech synthesis stress syllable synthesis Table techniques Technology tion training data transform triphones unigram University utterances variance vector quantization veriﬁcation vocabulary vocal tract voiced waveform word error rate