Weighted Viterbi algorithm and state duration modelling for speech recognition in noise.
Néstor Becerra YomaFergus R. McInnesMervyn A. JackPublished in: ICASSP (1998)
Keyphrases
- speech recognition
- hidden markov models
- viterbi algorithm
- noisy environments
- language model
- n gram
- automatic speech recognition
- speech recognizer
- speech processing
- speech signal
- speech synthesis
- noisy speech
- pattern recognition
- speech recognition technology
- state space
- speaker identification
- background noise
- speech retrieval
- dynamic programming
- speech recognizers
- isolated word
- information retrieval
- audio visual speech recognition
- speaker independent
- speech recognition systems
- markov model
- signal to noise ratio
- markov random field
- artificial neural networks
- reinforcement learning