Adaptation to a speaker's voice in a speech recognition system based on synthetic phoneme references.
Mats BlombergPublished in: Speech Commun. (1991)
Keyphrases
- speaker adaptation
- speech recognition
- prosodic features
- speech sounds
- speech synthesis
- automatic speech recognition
- speaker dependent
- text to speech
- maximum likelihood
- speaker verification
- vocal tract
- synthesized speech
- phoneme recognition
- speech signal
- real images are presented
- speaker independent
- automatic speech recognition systems
- hidden markov models
- audio visual
- voice activity detection
- speech recognizer
- speaker identification
- spontaneous speech
- neural network
- real world
- data sets
- visual speech
- adaptation strategies
- acoustic features
- adaptation process
- speaker recognition
- emotion recognition
- noisy environments
- multi modal
- language model
- machine learning