Improving Target Speaker Extraction with Sparse LDA-transformed Speaker Embeddings.
Kai LiuXucheng WanZiqing DuHuan ZhouPublished in: CoRR (2023)
Keyphrases
- speaker verification
- speech recognition
- speaker recognition
- dimensionality reduction
- audio visual
- face recognition
- speaker diarization
- automatic speech recognition
- feature extraction
- generative model
- latent dirichlet allocation
- speaker identification
- synthesized speech
- machine learning
- prosodic features
- acoustic features
- manifold learning
- discriminant analysis
- topic models
- image classification
- high dimensional