Recent advances in self-supervised speech models have shown significant improvement in many downstream tasks. However, these models predominantly centered on frame-level training objectives, which can fall short in spoken language understanding tasks that require semantic comprehension. Existing works often rely on additional speech-text data as intermediate targets, which is costly in the real-world setting. To address this challenge, we propose Pseudo-Word HuBERT (PW-HuBERT), a framework that integrates pseudo word-level targets into the training process, where the targets are derived from a visually-ground speech model, notably eliminating the need for speech-text paired data. Our experimental results on four spoken language understanding (SLU) benchmarks suggest the superiority of our model in capturing semantic information.
@article{arxiv.2402.05819,
title = {Integrating Self-supervised Speech Model with Pseudo Word-level Targets from Visually-grounded Speech Model},
author = {Hung-Chieh Fang and Nai-Xuan Ye and Yi-Jen Shih and Puyuan Peng and Hsuan-Fu Wang and Layne Berry and Hung-yi Lee and David Harwath},
journal= {arXiv preprint arXiv:2402.05819},
year = {2024}
}
Comments
Accepted to ICASSP 2024 workshop on Self-supervision in Audio, Speech, and Beyond (SASB)