VoxCeleb2: Deep Speaker Recognition

TLDR

The paper aims to advance speaker recognition in noisy, unconstrained settings by introducing a large‑scale dataset and developing CNN models. They built VoxCeleb2, a million‑utterance audio‑visual dataset from open‑source media, and trained CNNs with various strategies to recognize speakers. Models trained on VoxCeleb2 outperform prior work on benchmark data, and the dataset is several times larger than any publicly available speaker‑recognition dataset.

Abstract

The objective of this paper is speaker recognition under noisy and unconstrained conditions. We make two key contributions. First, we introduce a very large-scale audio-visual speaker recognition dataset collected from open-source media. Using a fully automated pipeline, we curate VoxCeleb2 which contains over a million utterances from over 6,000 speakers. This is several times larger than any publicly available speaker recognition dataset. Second, we develop and compare Convolutional Neural Network (CNN) models and training strategies that can effectively recognise identities from voice under various conditions. The models trained on the VoxCeleb2 dataset surpass the performance of previous works on a benchmark dataset by a significant margin.