ProsAudit, a prosodic benchmark for self-supervised speech models.
Maureen de SeysselMarvin LavechinHadrien TiteuxArthur ThomasGwendal VirletAndrea Santos RevillaGuillaume WisniewskiBogdan LudusanEmmanuel DupouxPublished in: CoRR (2023)
Keyphrases
- speech recognition
- real time
- database
- feature extraction
- comparative analysis
- neural network model
- statistical models
- experimental data
- complex systems
- text to speech synthesis
- acoustic models
- audio visual
- parameter estimation
- model selection
- multi modal
- probabilistic model
- prior knowledge
- artificial neural networks
- artificial intelligence