The evolving speech processing landscape is increasingly focused on complex scenarios like meetings or cocktail parties with multiple simultaneous speakers and far-field conditions. Existing methodologies for addressing these challenges fall into two categories: multi-channel and single-channel solutions. Single-channel approaches, notable for their generality and convenience, do not require specific information about microphone arrays. This paper presents a large-scale far-field overlapping speech dataset, crafted to advance research in speech separation, recognition, and speaker diarization. This dataset is a critical resource for decoding ``Who said What and When'' in multi-talker, reverberant environments, a daunting challenge in the field. Additionally, we introduce a pipeline system encompassing speech separation, recognition, and diarization as a foundational benchmark. Evaluations on the WHAMR! dataset validate the broad applicability of the proposed data.
@article{arxiv.2409.00819,
title = {LibriheavyMix: A 20,000-Hour Dataset for Single-Channel Reverberant Multi-Talker Speech Separation, ASR and Speaker Diarization},
author = {Zengrui Jin and Yifan Yang and Mohan Shi and Wei Kang and Xiaoyu Yang and Zengwei Yao and Fangjun Kuang and Liyong Guo and Lingwei Meng and Long Lin and Yong Xu and Shi-Xiong Zhang and Daniel Povey},
journal= {arXiv preprint arXiv:2409.00819},
year = {2024}
}