Speech Resynthesis from Discrete Disentangled Self-Supervised Representations.
Adam PolyakYossi AdiJade CopetEugene KharitonovKushal LakhotiaWei-Ning HsuAbdelrahman MohamedEmmanuel DupouxPublished in: CoRR (2021)
Keyphrases
- speech recognition
- speech signal
- discrete version
- symbolic representation
- automatic speech recognition systems
- data sets
- spoken language
- audio visual
- probabilistic model
- feature vectors
- pattern recognition
- discrete space
- speech processing
- discrete geometry
- speech synthesis
- distributed representations
- video sequences
- high level