Decoupling Segmental and Prosodic Cues of Non-native Speech through Vector Quantization.
Waris QuamerAnurag DasRicardo Gutierrez-OsunaPublished in: INTERSPEECH (2023)
Keyphrases
- vector quantization
- prosodic features
- speech synthesis
- text to speech
- speech recognition
- speaker verification
- text to speech synthesis
- speaker recognition
- image compression
- hidden markov models
- vector quantizer
- spontaneous speech
- reduced complexity
- fractal image compression
- input vector
- noisy environments
- distortion measure
- fractal image coding
- image processing
- codebook design
- finite state vector quantization
- automatic speech recognition
- speech signal
- neural gas
- emotion recognition
- audio visual
- synthesized speech
- motion estimation
- feature selection
- joint source and channel coding