Logo image
A 16nm 140TOPS/W 5&null Keyword Spotting Engine Based on 1D-BCNN
期刊文章   同儕審查

A 16nm 140TOPS/W 5&null Keyword Spotting Engine Based on 1D-BCNN

Tay-Jyi Lin, Yi-Hsuan Ting, Meng-Ze Hsu, Kuan-Han Lin, Chung-Ming Huang, Fu-Cheng Tsai, Shyh-Shyuan Sheu, Shih-Chieh Chang, Chingwei YehJinn-Shyan Wang
IEEE Transactions on Circuits and Systems II: Express Briefs
2023

摘要

1-dimensional binarized convolutional neural network Clocks Convolution Convolutional neural networks Current measurement energy efficient Engines Keyword spotting low power Random access memory Speech processing Electrical and Electronic Engineering
This brief presents an event-driven keyword spotting (KWS) system for reducing the significant but usually ignored energy dissipations on the &null A/D converter and microphone. &null energy per inference&null and &null responsiveness&null are new design goals of such KWS engine. A 7-layer 1-dimensional binarized convolutional neural network (1D-BCNN) was designed to achieve 95% inference accuracy for detecting 10 keywords, plus silence and unknown, from raw speech, and 64 32-element signed binary inner product units were allocated in the engine to deliver the 4,096 operations/cycle maximum throughput. The 16nm implementation consumes only 0.1mm2 silicon area and 5&null energy (including memory accesses), while achieving 1.72ms response time. The performance is comparable to state-of-the-art KWS designs without sacrificing number of detectable keywords or inference accuracy.

相關連結

指標

1 檢視次數

詳細資料

Logo image