arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.08057eess.AScs.AIcs.CLcs.SD
收藏

通过条件计算优化语音多视图特征融合

Optimizing Speech Multi-View Feature Fusion through Conditional Computation

Northeastern University(东北大学);The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳));Meituan(美团);Huawei Translation Services Center(华为翻译服务中心);NiuTrans Research(NiuTrans研究中心)

Weiqiao Shan, Yuhao Zhang, Yuchen Han, Bei Li, Xiaofeng Zhao, Yuang Li, Min Zhang, Hao Yang, Tong Xiao, Jingbo Zhu

AI总结:

针对自监督学习特征与传统频谱特征更新方向冲突的问题,提出带梯度敏感门控网络和多阶段dropout策略的条件计算特征融合框架,在MUSTC语音翻译任务上实现收敛加速且性能相当。

摘要:

近期研究进展凸显了自监督学习(self-supervised learning,SSL)特征在各类语音相关任务中的有效性,可提供轻量且通用的多视图语音表示。但本研究发现,SSL特征虽能加速模型收敛,却与FBank等传统频谱特征在更新方向上存在冲突。对此,我们提出一种基于条件计算的新型通用特征融合框架,配备梯度敏感门控网络与多阶段dropout策略。该框架可缓解特征冲突,增强模型对多视图输入特征的鲁棒性。通过融合SSL与频谱特征,我们的方法在MUSTC数据集的多项语音翻译任务上,既加速了收敛,又保持了与频谱模型相当的性能。

英文摘要:

Recent advancements have highlighted the efficacy of self-supervised learning (SSL) features in various speech-related tasks, providing lightweight and versatile multi-view speech representations. However, our study reveals that while SSL features expedite model convergence, they conflict with traditional spectral features like FBanks in terms of update directions. In response, we propose a novel generalized feature fusion framework grounded in conditional computation, featuring a gradient-sensitive gating network and a multi-stage dropout strategy. This framework mitigates feature conflicts and bolsters model robustness to multi-view input features. By integrating SSL and spectral features, our approach accelerates convergence and maintains performance on par with spectral models across multiple speech translation tasks on the MUSTC dataset.

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。

↑