DSVAE: Interpretable Disentangled Representation for Synthetic Speech Detection.
Amit Kumar Singh YadavKratika BhagtaniZiyue XiangPaolo BestaginiStefano TubaroEdward J. DelpPublished in: CoRR (2023)
Keyphrases
- speech recognition
- detection algorithm
- detection accuracy
- automatic detection
- false positives
- real world
- detection method
- computer vision
- spoken language
- audio visual
- image representation
- object detection
- multi modal
- detection rate
- image features
- feature representation
- knowledge base
- feature selection
- speaker recognition