Maximum mutual information based acoustic-features representation of phonological features for speech recognition.
Mohamed Kamal OmarMark Hasegawa-JohnsonPublished in: ICASSP (2002)
Keyphrases
- speech recognition
- automatic speech recognition
- speech signal
- mel frequency cepstral coefficients
- acoustic features
- speech recognition systems
- speaker identification
- hidden markov models
- language model
- cepstral coefficients
- noisy environments
- feature set
- feature extraction
- pattern recognition
- word recognition
- machine learning
- mutual information
- low level
- feature vectors
- gaussian mixture model
- audio features
- structural features
- noise reduction
- image features
- classification accuracy
- image processing