Warp-Q: Quality Prediction for Generative Neural Speech Codecs.
Wissam A. JassimJan SkoglundMichael ChinenAndrew HinesPublished in: ICASSP (2021)
Keyphrases
- quality prediction
- image quality
- speech recognition
- network architecture
- low complexity
- generative model
- neural network
- speech signal
- feature selection
- audio visual
- speech synthesis
- text to speech
- automatic speech recognition
- endpoint detection
- bio inspired
- data driven
- biologically inspired
- video codec
- language acquisition
- spoken language
- speaker recognition
- feature extraction
- learning rules
- visual quality
- neural model
- spoken dialogue systems
- vector quantization
- spike trains
- unsupervised learning
- recognition engine
- audio stream