2 June 2020 in IEEE Transactions on Circuits and Systems I: Regular Papers
This paper proposes a high power-performance-area efficient background noise aware keyword-spotting (KWS) processor based on an optimized binarized weight network (BWN). To reduce the power consumption while maintaining the system recognition accuracy for different background noise, the KWS processor with a SNR prediction module can be adaptively configured to use dual computing modes (standard computing mode and approximate computing mode) for both high recognition accuracy under high background noise and ultra-low power consumption under low background noise. The mel-scale frequency cepstral coefficients (MFCC) module is optimized with approximate computing technologies, which can reduce the power consumption by up to 3.1x and 5.7x for high/low background noise, respectively. Based on the evaluation of the architecture design space exploration, an ultra-low power BWN accelerator with low voltage, area and leakage power and using precision self-adaptive approximate computing units was proposed. Evaluated under 22nm process technology, this work can support up to 10 keywords real time recognition with power consumption of 15.1μW for high background noise and 10.8μW for low background noise. Compared to the state-of-the-art KWS architectures, our work can achieve ultra-low power consumption (about 1.7x reduced), while maintaining high system capability and adaptability.
21 December 2021 in IEEE Transactions on Circuits and Systems I: Regular Papers
Low-power keywords recognition has been a focus of acoustic signal processing for several decades. This work investigates the domain-specific speech recognition microprocessor based on optimized one-dimensional convolutional recurrent neural network (1D-CRNN). Compared to previous DNN based frameworks, the proposed 1D-CRNN can process both the feature extraction and keywords classification, and achieve high recognition accuracy with reduced computation operations under wide range background noise SNRs. An energy-efficient 1D-CRNN accelerator is implemented to dynamically reconfigure and process the different layers. This accelerator has the characteristics of “More is Less” in three aspects: 1) the hybrid network with more complex layers is much more compact and requires less computation; 2) although the weight width quantized to 8 bits requires more memory size and multiplication energy cost, the required network neurons can be reduced and hardware utilization can be improved; 3) an energy-aware self-compensation tensor multiplication unit with dual power supply based on approximation design method can be utilized for 1D-CRNN computing. Compared to the state-of-the-art architectures, the novel more-is-less architecture can achieve a much lower power consumption of $1.4~\mu \text{W}\sim 2.1~\mu \text{W}$ (over 80% reduced) under an industry 22nm technology, while maintaining higher system adaptability (support SNRs: −5dB~Clean) for 1~5 real-time keywords recognition.
19 December 2019 in IEEE Access
1 January 2017 in IEICE Electronics Express
This paper proposes an Energy-Efficient Reconfigurable Architecture (E-ERA) for Recurrent Neural Networks (RNNs). In E-ERA, reconfigurable computing arrays with approximate multipliers and dynamically adaptive accuracy controlling mechanism are implemented to achieve high energy efficiency. The E-ERA prototype is implemented on TSMC 45 nm process. Experimental results show that, comparing with traditional designs, the power consumption of E-ERA is reduced by 28.6%∼52.3%, with only 5.3%∼9.2% loss in accuracy. Compared with state-of-the-art architectures, E-ERA outperforms up to 1.78X in power efficiency and can achieve 304 GOPS/W when processing RNNs for speech recognition.
16 November 2020 in IEEE Access
The keyword spotting (KWS) system is one of the most important interfaces between humans and machines since it is usually the start of automatic speech recognition and natural language processing techniques. However, for KWS hardware, it is still a problem to make one specified chip both low power and high performed under multiple scenarios, such as in meeting rooms, on different traffic or in parks and so on, for different scenarios own wide range signal-noise-ratios (SNRs). The problem leads to the requirements of balanced design between KWS system accuracy and the hardware cost under various noise types and levels. To overcome the balanced design and tradeoff problems, a complete KWS processor including an Mel-Frequency Cepstrum Coefficients (MFCC) feature extractor and a quantized Convolutional Neural Network (QCNN) accelerator is proposed for wide SNR range and low-power KWS in this paper. Firstly, the approach to quantize CNNs into QCNNs with high accuracy is proposed with considerations of hardware-software tradeoff. With the tradeoff of KWS system accuracy and hardware cost, the 4bit/8bit dual-working-mode strategy is proposed to keep low hardware cost and high accuracy under different scenarios. To be specific, the training, tuning and validating of the CNNs and QCNNs are taken with the dataset of 10 keywords chosen from the Google Command Speech Dataset (GCSD). Secondly, a serial FFT based MFCC extractor is implemented with low power and small footprint. Finally, with a novel hybrid reuse strategy of input data and network weight, a reconfigurable and approximate computing based QCNN accelerator is designed. Implemented and verified under TSMC 22nm ULL technology, with the area of $1.42mm^{2}$ , the QCNN accelerator can achieve $5.26~\mu \text{W}/9.08~\mu \text{W}$ power consumption in 4bit/8bit work mode with accuracy of 88% and 93% respectively, which is superior to the state-of-the-art processors.
6 May 2025 in IEEE Transactions on Circuits and Systems II: Express Briefs
As technology node continues to shrink, the sensing margin of STT-MRAM is significantly diminished due to the downscaled supply voltage and increased process variations, posing challenges for sensing reliability. In this paper, a dual-swing-sample-and-couple sense amplifier is proposed to enhance the sensing margin and improve the read yield. By combining the data cell with two separate reference cells for sampling and coupling operations, the sensing margin is effectively doubled without pre-amplification. The simulation results demonstrate that the proposed circuit achieves a sensing margin up to 1.83 times that of conventional circuits, with a read yield of 99.9 at a sensing time of 2 ns. The read yield is not degraded at high temperatures (100 ∘ C), demonstrating high robustness. In addition, the extra power consumption induced by the preamplification operation is eliminated, effectively saving energy.