TsetlinKWS: A 65nm 16.58uW, 0.63mm2 State-Driven Convolutional Tsetlin Machine-Based Accelerator For Keyword Spotting
Organizations: University of Southampton, SO17 1BJ Southampton, U.K. · University College London, WC1E 6BT London, U.K.
Abstract
The Tsetlin Machine (TM) has recently attracted attention as a low-power alternative to neural networks due to its simple and interpretable inference mechanisms. However, its performance on speech-related tasks remains limited. This paper proposes TsetlinKWS, the first algorithm-hardware co-design framework for the Convolutional Tsetlin Machine (CTM) on the 12-keyword spotting task. Firstly, we introduce a novel Mel-Frequency Spectral Coefficient and Spectral Flux (MFSC-SF) feature extraction scheme together with spectral convolution, achieving a record-high accuracy of 87.35% for TM-based models on the 12-keyword spotting task. Secondly, we develop an Optimized Grouped Block-Compressed Sparse Row (OG-BCSR) algorithm that delivers a remarkable 9.84 reduction in model size, significantly improving the storage efficiency of CTMs. Finally, we propose a state-driven architecture tailored for CTMs, which simultaneously exploits data reuse and sparsity to attain high energy efficiency. The full system is evaluated in 65 nm process technology, consuming 16.58 W at 0.7 V with a compact 0.63 mm core area. TsetlinKWS demonstrates a Power-Performance-Area (PPA) product of 60.2 and requires only 907k logic operations per inference, corresponding to 1.4 and 10 reductions, respectively, compared with state-of-the-art KWS accelerators. These results position the CTM as a highly efficient candidate for ultra-low-power speech applications.
Figures & tables
| Batch-based Accelerator | Frame-based Accelerator | ||||
| This work | OJSSCS’23 [ 9 ] | TCAS-I’26 [ 13 ] | JSSC’20 [ 15 ] | TCASAI’24 [ 16 ] | |
| Technology | 65 nm | 22 nm | 28 nm | 65 nm | 65 nm |
| Voltage (V) | 0.7 | 0.54 | 0.42 | 0.6 | 0.6/0.65 |
| Core Area (mm 2 ) | 0.63 | ||||
| (FEx: 50%, Accel: 50%) | 0.6 | ||||
| (FEx: 42%, Accel: 58%) | 0.37 |
| Kernel Size | #Clause/Class | Training Accuracy @20-40dB | Test Accuracy @clean |
|---|---|---|---|
| 240 | 85.73% | 86.55% | |
| 120 | 85.33% | 87.35% | |
| 60 | 82.83% | 84.89% |
Keyword Spotting
Keyword spotting detects specified words or wake phrases in continuous audio, often on-device under tight latency, power, and false-alarm limits.
See all 27 papers