Label Aware Speech Representation Learning For Language Identification
Abstract
Speech representation learning approaches for non-semantic tasks such as language recognition have either explored supervised embedding extraction methods using a classifier model or self-supervised representation learning approaches using raw data. In this paper, we propose a novel framework of combining self-supervised representation learning with the language label information for the pre-training task. This framework, termed as Label Aware Speech Representation (LASR) learning, uses a triplet based objective function to incorporate language labels along with the self-supervised loss function. The speech representations are further fine-tuned for the downstream task. The language recognition experiments are performed on two public datasets - FLEURS and Dhwani. In these experiments, we illustrate that the proposed LASR framework improves over the state-of-the-art systems on language identification. We also report an analysis of the robustness of LASR approach to noisy/missing labels as well as its application to multi-lingual speech recognition tasks.
Cite
@article{arxiv.2306.04374,
title = {Label Aware Speech Representation Learning For Language Identification},
author = {Shikhar Vashishth and Shikhar Bharadwaj and Sriram Ganapathy and Ankur Bapna and Min Ma and Wei Han and Vera Axelrod and Partha Talukdar},
journal= {arXiv preprint arXiv:2306.04374},
year = {2023}
}
Comments
Accepted at Interspeech 2023