Towards Phonetically-Driven Hidden Markov Models: Can We Incorporate Phonetic Landmarks in HMM-Based ASR?
Guillaume GravierDaniel MoraruPublished in: NOLISP (2007)
Keyphrases
- hidden markov models
- speech recognition
- automatic speech recognition
- speech signal
- speaker independent
- speech recognizer
- gesture recognition
- noisy environments
- markov models
- conditional random fields
- sequential data
- sequence classification
- hidden state
- phoneme recognition
- keyword spotting
- hidden states
- visual speech recognition
- handwritten word recognition
- maximum likelihood
- discriminative training
- speaker identification
- broadcast news
- acoustic models
- handwriting recognition
- speaker adaptation
- markov model
- hierarchical hidden markov model