A unified maximum likelihood approach to acoustic mismatch compensation: application to noisy Lombard speech recognition.
Mohamed AfifyYifan GongJean-Paul HatonPublished in: ICASSP (1997)
Keyphrases
- speech recognition
- speech recognition systems
- noisy environments
- maximum likelihood
- language model
- hidden markov models
- pattern recognition
- speech processing
- speaker independent
- speech recognizers
- speech synthesis
- speaker identification
- speaker adaptation
- automatic speech recognition
- speech recognizer
- speech recognition technology
- speech retrieval
- audio visual speech recognition
- speech signal
- speech understanding
- acoustic models
- noisy speech
- neural network
- speaker verification
- feature extraction
- machine learning