Cascaded Multilingual Audio-Visual Learning from Videos

Abstract

In this paper, we explore self-supervised audio-visual models that learn from instructional videos. Prior work has shown that these models can relate spoken words and sounds to visual content after training on a large-scale dataset of videos, but they were only trained and evaluated on videos in English. To learn multilingual audio-visual representations, we propose a cascaded approach that leverages a model trained on English videos and applies it to audio-visual data in other languages, such as Japanese videos. With our cascaded approach, we show an improvement in retrieval performance of nearly 10x compared to training on the Japanese videos solely. We also apply the model trained on English videos to Japanese and Hindi spoken captions of images, achieving state-of-the-art performance.

arXiv commentPresented at Interspeech 2021. This version contains updated results using the YouCook-Japanese dataset

References in corpus (6)

• HowTo100M: Learning a Text-Video Embedding by Watching Hundred Million Narrated Video Clips
• Towards Automatic Learning of Procedures from Web Instructional Videos
• Representations of language in a model of visually grounded speech signal
• Language learning using Speech to Image retrieval
• Visually grounded models of spoken language: A survey of datasets, architectures and evaluation techniques
• Talk, Don't Write: A Study of Direct Speech-Based Image Retrieval