Token2vec: A Joint Self-Supervised Pre-Training Framework Using Unpaired Speech and Text-Reference-Cited by-同舟云学术

Token2vec: A Joint Self-Supervised Pre-Training Framework Using Unpaired Speech and Text

Published:2023-06-04 Issue: Volume: Page:
ISSN:
Container-title:ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
language:
Short-container-title:

Author:

Yue Xianghu¹,Ao Junyi²,Gao Xiaoxue¹,Li Haizhou¹

Affiliation:

1. National University of Singapore,Department of Electrical and Computer Engineering,Singapore

2. Shenzhen Research Institute of Big Data,Shenzhen,China

Funder

National Natural Science Foundation of China

Shenzhen Research Institute of Big Data

Chinese University of Hong Kong

Publisher

IEEE

Link

Reference33 articles.

1. Xtreme: A massively multilingual multi-task benchmark for evaluating cross-lingual generalisation;hu;ICML,2020

2. SuperGLUE: A stickier benchmark for general-purpose language understanding systems;wang;NeurIPS,2019

3. SLAM: A unified encoder for speech and language modeling via speechtext joint pre-training;bapna,2021

4. Bert: Pre-training of deep bidirectional transformers for language understanding;devlin;NAACL,2019

5. Librilight: A benchmark for asr with limited or no supervision;kahn;ICASSP,2020

Cited by 5 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

2. FastInject: Injecting Unpaired Text Data into CTC-Based ASR Training;ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP);2024-04-14

5. Few-Shot Spoken Language Understanding Via Joint Speech-Text Models;2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU);2023-12-16