Alpha-nets: a recurrent “neural” network architecture with a hidden Markov model interpretation
Speech Communication - Neurospeech
Connectionist Speech Recognition: A Hybrid Approach
Connectionist Speech Recognition: A Hybrid Approach
Hidden Markov Models for Speech Recognition
Hidden Markov Models for Speech Recognition
A novel connectionist-oriented feature normalization technique
ICANN'06 Proceedings of the 16th international conference on Artificial Neural Networks - Volume Part II
| Hi-index | 0.00 |
This paper introduces a novel combination of Artificial Neural Networks (ANNs) and Hidden Markov Models (HMMs) for Automatic SpeechRecognition (ASR), relying on ANN non-parametric estimation of the emission probabilities of an underlying HMM. A gradientascent global training technique aimed at maximizing the likelihood (ML) of acoustic observations given the model is presented. A maximum aposteriori variant of the algorithm is also proposed as a viable solution to the "divergence problem" that may arise in the ML setup. A 46.34% relative word error rate reduction withresp ect to standard HMMs was obtained in a speaker-independent, continuous ASR task witha small vocabulary.