LiCo-Net: Linearized Convolution Network for Hardware-efficient Keyword Spotting
This work addresses hardware efficiency for keyword spotting on microcontrollers, representing an incremental improvement over existing methods like SVDF.
The paper tackled the problem of hardware-efficient keyword spotting for low-power processors by proposing LiCo-Net, a dual-phase system using int8 linear operators at inference and streaming convolutions at training, which reduced cycles by 40% on HiFi4 DSP compared to SVDF while maintaining on-par detection performance.
This paper proposes a hardware-efficient architecture, Linearized Convolution Network (LiCo-Net) for keyword spotting. It is optimized specifically for low-power processor units like microcontrollers. ML operators exhibit heterogeneous efficiency profiles on power-efficient hardware. Given the exact theoretical computation cost, int8 operators are more computation-effective than float operators, and linear layers are often more efficient than other layers. The proposed LiCo-Net is a dual-phase system that uses the efficient int8 linear operators at the inference phase and applies streaming convolutions at the training phase to maintain a high model capacity. The experimental results show that LiCo-Net outperforms single-value decomposition filter (SVDF) on hardware efficiency with on-par detection performance. Compared to SVDF, LiCo-Net reduces cycles by 40% on HiFi4 DSP.