Improving body transmitted unvoiced speech with statistical voice conversion.
Mikihiro NakagiriTomoki TodaHideki KashiokaKiyohiro ShikanoPublished in: INTERSPEECH (2006)
Keyphrases
- speech quality
- speech signal
- text to speech
- speech recognition
- speech sounds
- speech synthesis
- fundamental frequency
- spectral analysis
- packet loss
- speech recognition errors
- background noise
- statistical analysis
- vocal tract
- statistical information
- data driven
- linear prediction
- automatic speech recognition
- statistical tests
- information theoretic
- statistical models
- non stationary
- neural network
- emotion recognition
- noisy environments
- speaker recognition
- progressive image transmission
- human body
- synthesized speech
- text to speech synthesis
- voice activity detection