Automatic speech recognition with sparse training data for dysarthric speakers.
Phil D. GreenJames CarmichaelAthanassios HatzisPam EnderbyMark S. HawleyMark ParkerPublished in: INTERSPEECH (2003)
Keyphrases
- automatic speech recognition
- speech recognition
- training data
- acoustic models
- hidden markov models
- speech signal
- speaker adaptation
- word error rate
- language model
- broadcast news
- spoken words
- speech retrieval
- decision trees
- conversational speech
- learning algorithm
- noisy environments
- speech synthesis
- speech corpus
- sparse representation
- speaker independent
- recognition errors
- formant frequencies
- word recognition
- training process
- high dimensional
- spontaneous speech
- speech recognizer
- acoustic features
- handwriting recognition
- text categorization
- multi modal
- pattern recognition