An HMM approach for synthesizing amused speech with a controllable intensity of smile.
Kevin El HaddadHüseyin ÇakmakAlexis MoinetStéphane DupontThierry DutoitPublished in: ISSPIT (2015)
Keyphrases
- speech recognition
- speech signal
- hidden markov models
- multi stream
- automatic speech recognition systems
- automatic speech recognition
- keyword spotting
- speech synthesis
- noisy speech
- speech processing
- phoneme recognition
- noisy environments
- audio visual
- handwritten word recognition
- speaker independent
- intensity values
- visual speech
- speech recognizer
- language model
- text to speech
- finite state transducers
- vocal tract
- speaker adaptation
- endpoint detection
- data sets
- language acquisition
- facial expression recognition
- non stationary
- pattern recognition
- neural network
- english text
- speaker recognition
- speech enhancement
- speaker identification
- finite state
- intensity levels
- maximum likelihood