Integer-only Zero-shot Quantization for Efficient Speech Recognition
This work addresses efficient deployment of ASR models on edge devices, offering a practical solution for scenarios with data privacy constraints, though it is incremental in improving quantization methods.
The paper tackles the problem of deploying large automatic speech recognition models on edge hardware by proposing an integer-only, zero-shot quantization scheme that eliminates the need for real data, achieving up to 2.35x speedup and 4x compression with less than 1% word error rate degradation.
End-to-end neural network models achieve improved performance on various automatic speech recognition (ASR) tasks. However, these models perform poorly on edge hardware due to large memory and computation requirements. While quantizing model weights and/or activations to low-precision can be a promising solution, previous research on quantizing ASR models is limited. In particular, the previous approaches use floating-point arithmetic during inference and thus they cannot fully exploit efficient integer processing units. Moreover, they require training and/or validation data during quantization, which may not be available due to security or privacy concerns. To address these limitations, we propose an integer-only, zero-shot quantization scheme for ASR models. In particular, we generate synthetic data whose runtime statistics resemble the real data, and we use it to calibrate models during quantization. We apply our method to quantize QuartzNet, Jasper, and Conformer and show negligible WER degradation as compared to the full-precision baseline models, even without using any data. Moreover, we achieve up to 2.35x speedup on a T4 GPU and 4x compression rate, with a modest WER degradation of <1% with INT8 quantization.