An energy-efficient voice activity detector using deep neural networks and approximate computing

Abstract This paper proposed an energy-efficient reconfigurable DNN accelerator architecture for voice activity detection (VAD) based on deep neural networks and fabricated in 28-nm technology. To reduce the power consumption and achieve high energy efficiency, two optimization techniques are proposed. First, the processing elements contained in the DNN accelerator support digital-analog mixed approximate computing, including multi-step quantized multiplication units and time-delay based addition units. Second, the proposed approximate computing units can be dynamically reconfigured to adapt to different computing accuracy requirements. The proposed approximate computing can significantly reduce the power consumption by 76% ∼ 88% compared to standard digital computing units. Implemented under TSMC 28 nm HPC + process technology, the layout size of the prototype system is 0.52 mm2, and the estimated power is 6 ∼ 12 μW. The energy efficiency of our work achieves 33.33 ∼ 66.67 TOPS/W, which is over 6.5X better than the state-of-the-art architecture.

[1]  James Tschanz,et al.  A 2.3 nJ/Frame Voice Activity Detector-Based Audio Front-End for Context-Aware System-On-Chip Applications in 32-nm CMOS , 2013, IEEE Journal of Solid-State Circuits.

[2]  Leibo Liu,et al.  A 141 UW, 2.46 PJ/Neuron Binarized Convolutional Neural Network Based Self-Learning Speech Recognition Processor in 28NM CMOS , 2018, 2018 IEEE Symposium on VLSI Circuits.

[3]  Yongqiang Wang,et al.  An investigation of deep neural networks for noise robust speech recognition , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[4]  Jia Wang,et al.  DaDianNao: A Machine-Learning Supercomputer , 2014, 2014 47th Annual IEEE/ACM International Symposium on Microarchitecture.

[5]  Janet M. Baker,et al.  The Design for the Wall Street Journal-based CSR Corpus , 1992, HLT.

[6]  Anantha P. Chandrakasan,et al.  A Low-Power Speech Recognizer and Voice Activity Detector Using Deep Neural Networks , 2018, IEEE Journal of Solid-State Circuits.

[7]  Leibo Liu,et al.  An Ultra-High Energy-Efficient Reconfigurable Processor for Deep Neural Networks with Binary/Ternary Weights in 28NM CMOS , 2018, 2018 IEEE Symposium on VLSI Circuits.

[8]  Leibo Liu,et al.  A High Energy Efficient Reconfigurable Hybrid Neural Network Processor for Deep Learning Applications , 2018, IEEE Journal of Solid-State Circuits.

[9]  Leibo Liu,et al.  A High Throughput Acceleration for Hybrid Neural Networks With Efficient Resource Management on FPGA , 2019, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[10]  Marian Verhelst,et al.  A 90 nm CMOS, $6\ {\upmu {\text{W}}}$ Power-Proportional Acoustic Sensing Frontend for Voice Activity Detection , 2016, IEEE Journal of Solid-State Circuits.

[11]  Marcin Pietras.  Error analysis in the hardware neural networks applications using reduced floating-point numbers representation , 2015 .

[12]  Leibo Liu,et al.  Deep Convolutional Neural Network Architecture With Reconfigurable Computation Patterns , 2017, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[13]  Eugenio Culurciello,et al.  Hardware accelerators for recurrent neural networks on FPGA , 2017, 2017 IEEE International Symposium on Circuits and Systems (ISCAS).

[14]  Wayne Luk,et al.  FP-BNN: Binarized neural network on FPGA , 2018, Neurocomputing.

[15]  Yu Wang,et al.  FPGA Acceleration of Recurrent Neural Network Based Language Model , 2015, 2015 IEEE 23rd Annual International Symposium on Field-Programmable Custom Computing Machines.

[16]  Ninghui Sun,et al.  DianNao: a small-footprint high-throughput accelerator for ubiquitous machine-learning , 2014, ASPLOS.

[17]  Vinod Kulathumani,et al.  Hibernets: Energy-Efficient Sensor Networks Using Analog Signal Processing , 2011, IEEE J. Emerg. Sel. Topics Circuits Syst..

[18]  Dong Yu,et al.  Context-Dependent Pre-Trained Deep Neural Networks for Large-Vocabulary Speech Recognition , 2012, IEEE Transactions on Audio, Speech, and Language Processing.

[19]  Yu Gong,et al.  EERA-ASR: An Energy-Efficient Reconfigurable Architecture for Automatic Speech Recognition With Hybrid DNN and Approximate Computing , 2018, IEEE Access.