A Multimode 157 μW 4-Channel 80 dBA-SNDR Speech Recognition Frontend With Direction-of-Arrival Correction Adaptive Beamformer
Taewook Kang, Seungjong Lee, Seungheun Song, Mohammad R. Haghighat, Michael P. Flynn · IEEE Journal of Solid-State Circuits · 2023
This work introduces a speech recognition frontend system that solves the problems of conventional adaptive beamforming (ABF) with: 1) low digital signal processing (DSP) power consumption (3$\times $lower than state-of-the-art ABF) thanks to an innovative greedy blocking matrix (GBM) employing simple calculations; 2) automatic direction-of-arrival (DOA) error compensation with direction tracking delay-and-sum beamformer aided by the GBM; 3) a multimode hybrid analog-to-digital converter (ADC) adapts to signal conditions; and 4) multimode beamforming takes advantage of high-signal signal-to-noise ratio (SNR) to reduce total power by 54%. A prototype fabricated in a 40 nm process occupies 0.94 mm2 while consuming 157 and 72$\mu \text{W}$in high-power and low-power modes, respectively. The proposed system improves speech recognition accuracy from 54% to 83% under noisy conditions.