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 2402.13076cs.SDcs.LGeess.AS
收藏

打破设备端流式语音识别中的功耗壁垒:洞察与解决方案

Breaking Down Power Barriers in On-Device Streaming ASR: Insights and Solutions

Meta;Iowa State University(爱荷华州立大学);Google(谷歌公司)

Yang Li, Yuan Shangguan, Yuhao Wang, Liangzhen Lai, Ernie Chang, Changsheng Zhao, Yangyang Shi, Vikas Chandra

AI总结:

本研究揭示语音识别模型权重参数配置对功耗的影响,提出按能量敏感性调整组件的设计原则,在保持精度的同时实现高达47%的能耗降低。

摘要:

功耗在设备端流式语音识别中起着至关重要的作用,显著影响用户体验。本研究探讨了语音识别模型中权重参数的配置如何影响其整体能效。我们发现,这些参数对功耗的影响因调用频率和内存分配等因素而异。利用这些洞察,我们提出了设计原则,通过在对精度影响最小的情况下降低功耗来增强设备端语音识别模型。我们的方法根据模型组件的特定能量敏感性进行调整,与领先方法相比,实现了高达47%的能耗降低,同时保持了相当的模型精度并提升了实时性能。

英文摘要:

Power consumption plays a crucial role in on-device streaming speech recognition, significantly influencing the user experience. This study explores how the configuration of weight parameters in speech recognition models affects their overall energy efficiency. We found that the influence of these parameters on power consumption varies depending on factors such as invocation frequency and memory allocation. Leveraging these insights, we propose design principles that enhance on-device speech recognition models by reducing power consumption with minimal impact on accuracy. Our approach, which adjusts model components based on their specific energy sensitivities, achieves up to 47% lower energy usage while preserving comparable model accuracy and improving real-time performance compared to leading methods.

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。

↑