2021 IEEE International Conference on Acoustics, Speech and Signal Processing
6-11 June 2021 • Toronto, Ontario, Canada
Extracting Knowledge from Information
| Paper ID | MLSP-40.4 | ||
| Paper Title | CONTRASTIVE LEARNING OF GENERAL-PURPOSE AUDIO REPRESENTATIONS | ||
| Authors | Aaqib Saeed, Eindhoven University of Technology, Netherlands; David Grangier, Neil Zeghidour, Google Research, France | ||
| Session | MLSP-40: Contrastive Learning | ||
| Location | Gather.Town | ||
| Session Time: | Friday, 11 June, 11:30 - 12:15 | ||
| Presentation Time: | Friday, 11 June, 11:30 - 12:15 | ||
| Presentation | Poster | ||
| Topic | Machine Learning for Signal Processing: [MLR-SSUP] Self-supervised and semi-supervised learning | ||
| IEEE Xplore Open Preview | Click here to view in IEEE Xplore | ||
| Abstract | We introduce COLA, a self-supervised pre-training approach for learning a general-purpose representation of audio. Our approach is based on contrastive learning: it learns a representation which assigns high similarity to audio segments extracted from the same recording while assigning lower similarity to segments from different recordings. We build on top of recent advances in contrastive learning for computer vision and reinforcement learning to design a lightweight, easy-to-implement self-supervised model of audio. We pre-train embeddings on the large-scale Audioset database and transfer these representations to 9 diverse classification tasks, including speech, music, animal sounds, and acoustic scenes. We show that despite its simplicity, our method significantly outperforms previous self-supervised systems. We furthermore conduct ablation studies to identify key design choices and release a library to pre-train and fine-tune COLA models. | ||