Unvoiced Speech Recognition Using Tissue-Conductive Acoustic Sensor.
Panikos HeracleousTomomi KainoHiroshi SaruwatariKiyohiro ShikanoPublished in: EURASIP J. Adv. Signal Process. (2007)
Keyphrases
- speech signal
- sound source
- acoustic features
- speech sounds
- speech recognition
- speech quality
- source localization
- automatic speech recognition
- speech recognition systems
- noisy speech
- formant frequencies
- vocal tract
- speaker identification
- emotional speech
- speaker independent
- speech recognizers
- background noise
- speech enhancement
- tissue classification
- sensor data
- acoustic models
- prosodic features
- spectral analysis
- non stationary
- sensor networks
- three dimensional
- image analysis
- sensor fusion
- visual speech
- noisy environments
- acoustic signal
- hidden markov models
- multi sensor
- speech segments
- text to speech
- mel frequency cepstral coefficients
- emotion recognition
- audio visual
- broadcast news
- recognition engine
- underwater vehicles
- vowel phonemes
- probabilistic model
- music information retrieval
- speech recognizer
- finite element model
- audio features
- contact force