Using KL-based acoustic models in a large vocabulary recognition task.
Guillermo AradillaHervé BourlardMathew Magimai-DossPublished in: INTERSPEECH (2008)
Keyphrases
- speaker independent
- speech recognition
- acoustic models
- continuous speech recognition
- hidden markov models
- sign language recognition
- speech recognizer
- pattern recognition
- speaker identification
- feature extraction
- noisy environments
- action recognition
- speech recognition systems
- digit recognition
- speech signal
- information retrieval
- semantic analysis
- character recognition
- image processing
- computer vision