arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2409.09408eess.AScs.SD
收藏

利用自监督学习进行说话人日志

Leveraging Self-Supervised Learning for Speaker Diarization

Brno University of Technology(布尔诺理工大学)

Jiangyu Han, Federico Landini, Johan Rohdin, Anna Silnova, Mireia Diez, Lukas Burget

AI总结:

本文利用WavLM自监督表示结合Conformer改进端到端神经说话人日志,在AMI、AISHELL-4等数据集上超越Pyannote基线并取得新SOTA,同时证明其对数据稀缺更具鲁棒性。

摘要:

端到端神经日志在过去几年中取得了长足发展,但数据稀缺仍然是进一步改进的主要障碍。诸如WavLM之类的自监督学习方法已在多个下游任务上展现出有前景的性能,但它们在说话人日志中的应用仍然较为有限。在本工作中,我们探索使用WavLM来缓解神经日志训练中的数据稀缺问题。我们采用与Pyannote相同的流程,并利用WavLM和Conformer改进局部端到端神经日志。在远场AMI、AISHELL-4和AliMeeting数据集上的实验表明,我们的方法显著优于Pyannote基线,并分别在AMI和AISHELL-4上取得了新的最先进结果。此外,通过分析不同数据量场景下的系统性能,我们表明WavLM表示对数据稀缺的鲁棒性远高于滤波器组特征,从而支持数据需求更低的训练策略。我们还发现,在我们的实验中,通常用于训练端到端日志模型的模拟数据在使用WavLM时并无帮助。此外,我们还在最近的CHiME8 NOTSOFAR-1任务上评估了我们的模型,其性能优于Pyannote基线。我们的源代码公开于https://github.com/BUTSpeechFIT/DiariZen。

英文摘要:

End-to-end neural diarization has evolved considerably over the past few years, but data scarcity is still a major obstacle for further improvements. Self-supervised learning methods such as WavLM have shown promising performance on several downstream tasks, but their application on speaker diarization is somehow limited. In this work, we explore using WavLM to alleviate the problem of data scarcity for neural diarization training. We use the same pipeline as Pyannote and improve the local end-to-end neural diarization with WavLM and Conformer. Experiments on far-field AMI, AISHELL-4, and AliMeeting datasets show that our method substantially outperforms the Pyannote baseline and achieves new state-of-the-art results on AMI and AISHELL-4, respectively. In addition, by analyzing the system performance under different data quantity scenarios, we show that WavLM representations are much more robust against data scarcity than filterbank features, enabling less data hungry training strategies. Furthermore, we found that simulated data, usually used to train endto-end diarization models, does not help when using WavLM in our experiments. Additionally, we also evaluate our model on the recent CHiME8 NOTSOFAR-1 task where it achieves better performance than the Pyannote baseline. Our source code is publicly available at https://github.com/BUTSpeechFIT/DiariZen.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑