Two-step unsupervised speaker adaptation based on speaker and gender recognition and HMM combination.
Petr CervaJan NouzaJan SilovskýPublished in: INTERSPEECH (2006)
Keyphrases
- speaker adaptation
- gender recognition
- speech recognition
- maximum likelihood
- automatic speech recognition
- speech recognizer
- speaker dependent
- hidden markov models
- speaker independent
- depth images
- gender classification
- speech signal
- semi supervised
- speaker identification
- face images
- probabilistic model
- pattern recognition
- machine learning
- language model