SMS-WSJ
Emerging9papers using it
2021first seen
The SMS-WSJ dataset/benchmark contains mixed audio recordings of multiple speakers and is used to evaluate the performance of speech recognition systems in far-field multi-speaker environments.
Papers using SMS-WSJ (9)
- Multi-channel multi-speaker transformer for speech recognitionNeural Forward Filtering for Speaker-Image SeparationVM-UNSSOR: Unsupervised Neural Speech Separation Enhanced by Higher-SNR Virtual Microphone ArraysElevating Robust Multi-Talker ASR by Decoupling Speaker Separation and
Speech RecognitionConvolutive Prediction for Monaural Speech Dereverberation and
Noisy-Reverberant Speaker SeparationConvolutive Prediction for Reverberant Speech SeparationMMS-MSG: A Multi-purpose Multi-Speaker Mixture Signal GeneratorTF-GridNet: Integrating Full- and Sub-Band Modeling for Speech
SeparationMixture Encoder for Joint Speech Separation and Recognition