An Effective Transformer-based Contextual Model and Temporal Gate Pooling for Speaker Identification

TL;DR

This paper presents a novel Transformer-based speaker identification model with a new pooling method, achieving high accuracy with fewer parameters compared to existing models, and explores hyper-parameter effects on performance.

Contribution

Introduces an end-to-end Transformer-based speaker identification model with Temporal Gate Pooling and analyzes hyper-parameter impacts on performance.

Findings

Achieved 87.1% accuracy on VoxCeleb1 with 28.5M parameters

Comparable precision to wav2vec2 with significantly fewer parameters

Proposed Temporal Gate Pooling enhances speaker identification performance

Abstract

Wav2vec2 has achieved success in applying Transformer architecture and self-supervised learning to speech recognition. Recently, these have come to be used not only for speech recognition but also for the entire speech processing. This paper introduces an effective end-to-end speaker identification model applied Transformer-based contextual model. We explored the relationship between the hyper-parameters and the performance in order to discern the structure of an effective model. Furthermore, we propose a pooling method, Temporal Gate Pooling, with powerful learning ability for speaker identification. We applied Conformer as encoder and BEST-RQ for pre-training and conducted an evaluation utilizing the speaker identification of VoxCeleb1. The proposed method has achieved an accuracy of 87.1% with 28.5M parameters, demonstrating comparable precision to wav2vec2 with 317.7M parameters. Code is available at https://github.com/HarunoriKawano/speaker-identification-with-tgp.