AVA-AVD: Audio-Visual Speaker Diarization in the Wild

TL;DR

This paper introduces the AVA-AVD dataset for in-the-wild audio-visual speaker diarization, along with a new model called AVR-Net that improves robustness and accuracy in challenging scenarios.

Contribution

The paper creates the first in-the-wild audio-visual diarization dataset and proposes AVR-Net, a novel model with a modality mask for better speaker discrimination.

Findings

Adding AVA-AVD improves diarization performance in wild videos.

AVR-Net outperforms state-of-the-art methods.

The model is more robust to off-screen speakers.

Abstract

Audio-visual speaker diarization aims at detecting "who spoke when" using both auditory and visual signals. Existing audio-visual diarization datasets are mainly focused on indoor environments like meeting rooms or news studios, which are quite different from in-the-wild videos in many scenarios such as movies, documentaries, and audience sitcoms. To develop diarization methods for these challenging videos, we create the AVA Audio-Visual Diarization (AVA-AVD) dataset. Our experiments demonstrate that adding AVA-AVD into training set can produce significantly better diarization models for in-the-wild videos despite that the data is relatively small. Moreover, this benchmark is challenging due to the diverse scenes, complicated acoustic conditions, and completely off-screen speakers. As a first step towards addressing the challenges, we design the Audio-Visual Relation Network (AVR-Net) which introduces a simple yet effective modality mask to capture discriminative information based on face visibility. Experiments show that our method not only can outperform state-of-the-art methods but is more robust as varying the ratio of off-screen speakers. Our data and code has been made publicly available at https://github.com/showlab/AVA-AVD.