Speech Recognition Using Acoustic Similarity-Based Primitives.
Takafumi HayashiHiroki MoriMotoyuki SuzukiShozo MakinoHirotomo AsoPublished in: Systems and Computers in Japan (2002)
Keyphrases
- speech sounds
- speech recognition
- speech recognition systems
- acoustic features
- prosodic features
- emotional speech
- speech signal
- speaker independent
- automatic speech recognition
- speech synthesis
- acoustic models
- speech recognizers
- noisy speech
- acoustic signal
- vocal tract
- building blocks
- speaker verification
- hidden markov models
- low level
- formant frequencies
- emotion recognition
- speech recognizer
- source localization
- vowel phonemes
- text to speech
- pattern recognition
- high level
- neural network
- speech enhancement
- visual speech
- spontaneous speech
- spectral features
- speaker identification
- linear prediction
- feature extraction
- automatic speech recognition systems
- data sets