WSJ

The WSJ dataset contains transcribed speech data and is used to evaluate the performance of speech recognition systems, particularly in the context of unsupervised domain adaptation.

Papers using WSJ (13)

Fearless: Feature Refinement Loss For Ensembling Self-supervised Learning Features In Robust End-to-end Speech Recognition2022 · 8 citesTowards Decoupling Frontend Enhancement And Backend Recognition In Monaural Robust ASR2024 · 3 citesMixRep: Hidden Representation Mixup for Low-Resource Speech Recognition2023 · 5 citesTeaching the Teachers: Boosting unsupervised domain adaptation in speech recognition by ensemble update2026Boosting CTC-Based ASR Using LLM-Based Intermediate Loss Regularization2025 · 1 citesGenerative Speech Recognition Error Correction with Large Language Models and Task-Activating Prompting2023 · 2 citesTowards Decoupling Frontend Enhancement and Backend Recognition in Monaural Robust ASR2024 · 1 citesInvestigation of Ensemble features of Self-Supervised Pretrained Models for Automatic Speech Recognition2022FeaRLESS: Feature Refinement Loss for Ensembling Self-Supervised Learning Features in Robust End-to-end Speech Recognition2022MMS-MSG: A Multi-purpose Multi-Speaker Mixture Signal Generator2022Minimum Latency Training of Sequence Transducers for Streaming End-to-End Speech Recognition2022Progressive unsupervised domain adaptation for ASR using ensemble models and multi-stage training2024Noise-robust Speech Separation with Fast Generative Correction2024