Between recognition and synthesis - 300 bits/second speech coding.
Mohamed IsmailKeith PontingPublished in: EUROSPEECH (1997)
Keyphrases
- recognition engine
- recognition rate
- recognition accuracy
- speech recognition
- object recognition
- speech corpus
- text recognition
- pattern recognition
- coding scheme
- gray code
- linear prediction
- speech signal
- coding method
- speech recognition systems
- speaker independent
- continuous speech recognition
- automatic transcription
- spoken words
- automatic speech recognition systems
- feature extraction
- speaker identification
- inter frame
- character recognition
- language model
- hidden markov models
- speech sounds
- significant bit