A Sidecar Separator Can Convert a Single-Speaker Speech Recognition System to a Multi-Speaker One.
Lingwei MengJiawen KangMingyu CuiYuejiao WangXixin WuHelen MengPublished in: CoRR (2023)
Keyphrases
- speech recognition
- speaker verification
- audio visual
- automatic speech recognition
- speaker recognition
- speaker dependent
- information retrieval
- artificial intelligence
- pattern recognition
- speaker identification
- speaker diarization
- prosodic features
- machine learning
- social networks
- evolutionary algorithm
- synthesized speech