Speech/Non-Speech Segmentation Based on Phoneme Recognition Features.
Janez ZibertNikola PavesicFrance MihelicPublished in: EURASIP J. Adv. Signal Process. (2006)
Keyphrases
- phoneme recognition
- speaker dependent
- hidden markov models
- audio signal
- speech recognition
- automatic speech recognition
- speech signal
- speech recognition systems
- spectral features
- feature extraction
- speaker identification
- spoken language
- broadcast news
- feature vectors
- extracting features
- audio visual
- image features
- speech synthesis
- speech corpus
- edge detection
- image segmentation
- low level
- speaker recognition
- classification accuracy
- feature set