16 search results - page 1 / 4 » High-Accuracy Large-Vocabulary Speech Recognition Using Mixt... |
NAACL
1994
13 years 6 months ago
1994
Improved acoustic modeling can significantly decrease the error rate in large-vocabulary speech recognition. Our approach to the problem is twofold. We first propose a scheme that...
NAACL
1994
13 years 6 months ago
1994
The key problem to be faced when building a HMM-based continuous speech recogniser is maintaining the balance between model complexity and available training data. For large vocab...
ICASSP
2011
IEEE
12 years 8 months ago
2011
IEEE
The context-independent deep belief network (DBN) hidden Markov model (HMM) hybrid architecture has recently achieved promising results for phone recognition. In this work, we pro...
ICASSP
2011
IEEE
12 years 8 months ago
2011
IEEE
Recently there has been interest in structured discriminative models for speech recognition. In these models sentence posteriors are directly modelled, given a set of features ext...
NAACL
2003
13 years 6 months ago
2003
It is well known that frame independence assumption is a fundamental limitation of current HMM based speech recognition systems. By treating each speech frame independently, HMMs ...