WSJ-0-2/3 Mix
Emerging5papers using it
2021first seen
The 'WSJ-0-2/3 Mix' dataset is a benchmark used to evaluate speech separation performance, containing mixed speech signals from the Wall Street Journal corpus.
Papers using WSJ-0-2/3 Mix (5)
- Exploring Self-Attention Mechanisms for Speech SeparationMossFormer: Pushing the Performance Limit of Monaural Speech Separation using Gated Single-Head Transformer with Convolution-Augmented Joint Self-AttentionsMossFormer2: Combining Transformer and RNN-Free Recurrent Network for Enhanced Time-Domain Monaural Speech SeparationTiny-Sepformer: A Tiny Time-Domain Transformer Network for Speech SeparationStepwise-Refining Speech Separation Network via Fine-Grained Encoding in High-order Latent Domain