# Towards Lightweight Speaker Verification via Adaptive Neural Network Quantization

**Authors:** Bei Liu, Haoyu Wang, Yanmin Qian  
**arXiv:** [2406.05359](https://arxiv.org/abs/2406.05359)  
**Date:** 2024-06-08  
**Summary page:** https://www.summarizepaper.com/en/paper/2406.05359v1/

## Key points

- Authors address the challenge of deploying modern speaker verification systems on mobile devices due to high demand for storage and computing resources
- Proposed approach: lightweight speaker verification through adaptive neural network quantization
- Development of an adaptive uniform precision quantization method using k-means clustering for dynamic generation of quantization centroids tailored to each network layer
- Introduction of mixed precision quantization algorithm and multi-stage fine-tuning strategy to improve performance of low-bit quantized models
- Design of two distinct binary quantization schemes (static and adaptive) to address performance degradation in 1-bit quantized models
- Experimental results show lossless 4-bit uniform precision quantization can be achieved with promising compression ratio, outperforming existing methods across various model size ranges

## Summary

In their paper titled "Towards Lightweight Speaker Verification via Adaptive Neural Network Quantization," authors Bei Liu, Haoyu Wang, and Yanmin Qian address the challenge of deploying modern speaker verification (SV) systems on mobile devices due to their high demand for storage and computing resources. The authors propose an innovative approach to lightweight speaker verification through adaptive neural network quantization. The  of this research lies in the development of an adaptive uniform precision quantization method that allows for the dynamic generation of quantization centroids tailored to each network layer using k-means clustering. By applying this method to pre-trained SV systems, the authors generate a series of quantized variants with different bit widths. To improve the performance of low-bit quantized models, they introduce a mixed precision quantization algorithm along with a multi-stage fine-tuning (MSFT) strategy. Unlike traditional uniform precision quantization methods, the mixed precision approach enables assigning varying bit widths to different network layers. Once the optimal bit combination is determined, MSFT is employed to progressively quantize and fine-tune the network in a specific order. Additionally, two distinct binary quantization schemes are designed to address performance degradation in 1-bit quantized models: static and adaptive quantizers. Experimental results on VoxCeleb demonstrate that lossless 4-bit uniform precision quantization can be achieved on both ResNets and DF-ResNets, leading to a promising compression ratio of around 8. Furthermore, compared to uniform precision methods, mixed precision quantization not only enhances performance with a similar model size but also offers flexibility in generating bit combinations for desired model sizes. The authors' proposed 1-bit  significantly improve the performance of binarized models. A comprehensive comparison with existing lightweight SV systems shows that the proposed models outperform previous methods by a significant margin across various model size ranges. This research has been submitted for review to IEEE/ACM Transactions on Audio Speech and Language Processing, showcasing its potential impact on advancing lightweight speaker verification technologies.