Reducing the corpus-based TTS signal degradation due to speaker's word pronunciations.
Sérgio PauloLuís C. OliveiraPublished in: INTERSPEECH (2005)
Keyphrases
- keywords
- prosodic features
- speech recognition
- text to speech
- speech synthesis
- speaker verification
- noisy environments
- speech recognition systems
- automatic speech recognition
- speech recognizer
- signal processing
- non stationary
- signal detection
- speaker identification
- language model
- pattern recognition
- high frequency
- frequency domain
- original signal
- spontaneous speech
- speech segments
- out of vocabulary
- word recognition
- co occurrence
- speaker recognition
- compressive sensing
- audio visual