Separating phonetic and speaker features of vowels in formant space.
Kung-Pu LiPublished in: ICASSP (1987)
Keyphrases
- speech recognition
- speech recognition systems
- speech signal
- speech recognizer
- automatic speech recognition
- language model
- mel frequency cepstral coefficients
- feature space
- image features
- speaker verification
- speaker identification
- speech synthesis
- noisy environments
- false positives
- speaker dependent
- feature vectors
- pattern recognition
- feature extraction
- acoustic models
- vocal tract
- data sets
- emotional speech
- extracting features
- emotion recognition
- cross correlation
- feature set
- classification accuracy
- low level
- hidden markov models
- prior knowledge