Integrating multiple pronunciations during MCE-based acoustic model training for large vocabulary speech recognition.
Rathi ChengalvarayanPublished in: INTERSPEECH (2002)
Keyphrases
- speech recognition
- integrating multiple
- wall street journal corpus
- speech recognizer
- hidden markov models
- isolated word
- minimum classification error
- speech recognition systems
- acoustic models
- language model
- speech synthesis
- pattern recognition
- speech processing
- automatic speech recognition
- speech signal
- speech recognition technology
- speaker independent
- training process
- feature space
- speaker identification
- speaker adaptation
- speech recognizers
- speech retrieval
- training set
- natural language processing
- bayesian networks
- speaker dependent
- discriminative training
- noisy environments