Multi-level context-dependent acoustic modeling for automatic speech recognition.
Hung-An ChangJames R. GlassPublished in: ASRU (2011)
Keyphrases
- context dependent
- automatic speech recognition
- acoustic features
- speech recognition
- speech sounds
- speech recognizers
- hidden markov models
- speech segments
- conversational speech
- low level
- speech signal
- spoken words
- broadcast news
- context free
- noisy environments
- acoustic models
- recognition errors
- word error rate
- natural language
- speech recognition systems
- speech corpus
- image processing
- maximum likelihood