Language-Based Audio Retrieval with Converging Tied Layers and Contrastive Loss
Sound
2022-06-30 v1 Computation and Language
Information Retrieval
Audio and Speech Processing
Abstract
In this paper, we tackle the new Language-Based Audio Retrieval task proposed in DCASE 2022. Firstly, we introduce a simple, scalable architecture which ties both the audio and text encoder together. Secondly, we show that using this architecture along with contrastive loss allows the model to significantly beat the performance of the baseline model. Finally, in addition to having an extremely low training memory requirement, we are able to use pretrained models as it is without needing to finetune them. We test our methods and show that using a combination of our methods beats the baseline scores significantly.
Cite
@article{arxiv.2206.14659,
title = {Language-Based Audio Retrieval with Converging Tied Layers and Contrastive Loss},
author = {Andrew Koh and Eng Siong Chng},
journal= {arXiv preprint arXiv:2206.14659},
year = {2022}
}