Speaker independent speech recognition method using constrained time alignment near phoneme discriminative frame.
Tomohiro KonumaTetsu SuzukiMaki YamadaYoshio OnoMasakatsu HoshimiKatsuyuki NiyadaPublished in: ICSLP (1998)
Keyphrases
- speech recognition
- hidden markov models
- speaker independent
- speech recognizer
- pattern recognition
- speech synthesis
- speaker adaptation
- noisy speech
- speaker dependent
- similarity measure
- automatic speech recognition
- language model
- speech recognition systems
- neural network
- noisy environments
- speech signal
- probabilistic model
- video sequences
- bayesian networks