当前位置: X-MOL 学术IEEE Trans. Circuits Syst. I Regul. Pap. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
NS-FDN: Near-Sensor Processing Architecture of Feature-Configurable Distributed Network for Beyond-Real-Time Always-on Keyword Spotting
IEEE Transactions on Circuits and Systems I: Regular Papers ( IF 5.2 ) Pub Date : 2021-02-24 , DOI: 10.1109/tcsi.2021.3059649
Qin Li , Changlu Liu , Peiyan Dong , Yanming Zhang , Tong Li , Sheng Lin , Minda Yang , Fei Qiao , Yanzhi Wang , Li Luo , Huazhong Yang

Always-on keyword spotting (KWS) that detects wake-up words has been the indispensable module in the voice interaction system. However, the ultra-low-power embedded devices put forward strict requirements on energy consumption, latency, and recognition accuracy of KWS. In this work, we propose a near-sensor processing architecture of feature-configurable distributed network (NS-FDN) for always-on KWS applications. The proposed distributed network adapts to the flexible keywords demands in the actual scene by splitting the conventional single network into distributed sub-networks. We design a channel-independent training framework to improve the recognition accuracy of distributed networks. The speech features are evaluated and the redundancy is reduced in NS-FDN, which can also configure the speech features to further reduce the computing complexity and improve processing speed. For deeper optimization, we implement a 65nm-process prototype chip with near-sensor mixed-signal processing architecture avoiding energy-consuming analog-to-digital converter. By improving the system, algorithm, and hardware designs of the KWS, our co-optimized architecture eliminates the energy consumption bottleneck long-standing in conventional KWS systems and achieves state-of-the-art system performance. The experiment results show that NS-FDN achieves 31.6% energy consumption savings, 1.6 times memory savings, 57 times speedup, and 3.4% higher recognition accuracy compared with the state of the art.

中文翻译:

NS-FDN:超出实时始终在线关键字识别功能的可配置分布式网络的近传感器处理体系结构

语音交互系统中必不可少的模块是检测唤醒单词的Always-on关键字搜寻(KWS)。但是,超低功耗嵌入式设备对KWS的能耗,等待时间和识别精度提出了严格的要求。在这项工作中,我们为始终在线的KWS应用提出了一种功能可配置的分布式网络(NS-FDN)的近传感器处理体系结构。所提出的分布式网络通过将常规的单个网络划分为分布式子网来适应实际场景中的灵活关键词需求。我们设计了与通道无关的训练框架,以提高分布式网络的识别准确性。评估语音功能并减少NS-FDN中的冗余,它还可以配置语音功能,以进一步降低计算复杂度并提高处理速度。为了进行更深入的优化,我们实现了具有65nm工艺的原型芯片,该芯片具有近传感器混合信号处理架构,从而避免了耗能的模数转换器。通过改进KWS的系统,算法和硬件设计,我们共同优化的体系结构消除了传统KWS系统中长期存在的能耗瓶颈,并实现了最新的系统性能。实验结果表明,与现有技术相比,NS-FDN节省了31.6%的能耗,节省了1.6倍的内存,加速了57倍,并提高了3.4%的识别精度。我们采用具有近传感器混合信号处理架构的65nm工艺原型芯片,避免了耗能的模数转换器。通过改进KWS的系统,算法和硬件设计,我们共同优化的体系结构消除了传统KWS系统中长期存在的能耗瓶颈,并实现了最新的系统性能。实验结果表明,与现有技术相比,NS-FDN节省了31.6%的能耗,节省了1.6倍的内存,加速了57倍,并提高了3.4%的识别精度。我们采用具有近传感器混合信号处理架构的65nm工艺原型芯片,避免了耗能的模数转换器。通过改进KWS的系统,算法和硬件设计,我们共同优化的体系结构消除了传统KWS系统中长期存在的能耗瓶颈,并实现了最新的系统性能。实验结果表明,与现有技术相比,NS-FDN节省了31.6%的能耗,节省了1.6倍的内存,加速了57倍,并提高了3.4%的识别精度。我们共同优化的架构消除了传统KWS系统长期以来的能耗瓶颈,并实现了最先进的系统性能。实验结果表明,与现有技术相比,NS-FDN节省了31.6%的能耗,节省了1.6倍的内存,加速了57倍,并提高了3.4%的识别精度。我们共同优化的架构消除了传统KWS系统长期以来的能耗瓶颈,并实现了最先进的系统性能。实验结果表明,与现有技术相比,NS-FDN节省了31.6%的能耗,节省了1.6倍的内存,加速了57倍,并提高了3.4%的识别精度。
更新日期:2021-04-20
down
wechat
bug