Open AccessEERA-ASR: An Energy-Efficient Reconfigurable Architecture for Automatic Speech Recognition With Hybrid DNN and Approximate Computing
Author(s) -
Bo Liu,
Hai Qin,
Yu Gong,
Wei Ge,
Mengwen Xia,
Longxing Shi
Publication year - 2018
Publication title -
ieee access
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.587
H-Index - 127
ISSN - 2169-3536
DOI - 10.1109/access.2018.2870273
Subject(s) - aerospace , bioengineering , communication, networking and broadcast technologies , components, circuits, devices and systems , computing and processing , engineered materials, dielectrics and plasmas , engineering profession , fields, waves and electromagnetics , general topics for engineers , geoscience , nuclear engineering , photonics and electrooptics , power, energy and industry applications , robotics and control systems , signal processing and analysis , transportation
This paper proposes a hybrid deep neural network (DNN) for automatic speech recognition and an energy-efficient reconfigurable architecture with approximate computing for accelerating the DNN. To accelerate the hybrid DNN and reduce the energy consumption, we propose a digital-analog mixed reconfigurable architecture with approximate computing units, including a binary weight network accelerator with analog multi-chain delay-addition units for bit-wise approximate computing and a recurrent neural network accelerator with approximate multiplication units for different calculation accuracy requirements. Implemented under TSMC 28nm HPC+ process technology, the proposed architecture can achieve the energy efficiency of 163.8TOPS/W for 20 keywords recognition and 3.3TOPS/W for common speech recognition.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom