MR-RawNet: Speaker verification system with multiple temporal resolutions for variable duration utterances using raw waveforms

TL;DR

This paper introduces MR-RawNet, a novel speaker verification system that leverages multi-resolution features and attention mechanisms to improve robustness across variable-length utterances using raw waveforms.

Contribution

The paper proposes a multi-resolution feature extractor and attention block that enhance speaker verification performance on variable-duration utterances from raw waveforms.

Findings

Outperforms existing raw waveform-based systems on VoxCeleb1

Improves robustness to utterance length variability

Effectively captures diverse temporal contexts

Abstract

In speaker verification systems, the utilization of short utterances presents a persistent challenge, leading to performance degradation primarily due to insufficient phonetic information to characterize the speakers. To overcome this obstacle, we propose a novel structure, MR-RawNet, designed to enhance the robustness of speaker verification systems against variable duration utterances using raw waveforms. The MR-RawNet extracts time-frequency representations from raw waveforms via a multi-resolution feature extractor that optimally adjusts both temporal and spectral resolutions simultaneously. Furthermore, we apply a multi-resolution attention block that focuses on diverse and extensive temporal contexts, ensuring robustness against changes in utterance length. The experimental results, conducted on VoxCeleb1 dataset, demonstrate that the MR-RawNet exhibits superior performance in handling utterances of variable duration compared to other raw waveform-based systems.