Multilingual Meta-Transfer Learning for Low-Resource Speech Recognition

This paper proposes a novel meta-transfer learning method to improve automatic speech recognition (ASR) performance in low-resource languages. Nowadays, we are witnessing high interest in low-resource ASR tasks aiming at delivering feasible and reliable systems with very limited data. The main chall...

Full description

Saved in:
Bibliographic Details
Published in:IEEE access 2024, Vol.12, p.158493-158504
Main Authors: Zhou, Rui, Koshikawa, Takaki, Ito, Akinori, Nose, Takashi, Chen, Chia-Ping
Format: Article
Language:English
Subjects:
Adaptation models
Automatic speech recognition
Corpus linguistics
Data models
Emotion recognition
end-to-end speech recognition
Error analysis
Error reduction
Knowledge transfer
Languages
low-resource languages
Machine learning
Meta-transfer learning
Metalearning
Multilingualism
Performance evaluation
Speech recognition
Sun
Training
Training data
Transfer learning
Voice recognition
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper proposes a novel meta-transfer learning method to improve automatic speech recognition (ASR) performance in low-resource languages. Nowadays, we are witnessing high interest in low-resource ASR tasks aiming at delivering feasible and reliable systems with very limited data. The main challenge is the design and development of a methodology to address the issue of data scarcity. Our proposed meta-transfer learning approach combines two well-known machine-learning methods: transfer learning and meta-learning. We propose their integration that can ameliorate the training bottlenecks and overfitting issues with pre-training models on low-resource speech data. For evaluation, we conduct extensive multilingual ASR experiments on the Common Voice Corpus and Globalphone Corpus and compare the performance of the meta-transfer learning, meta-learning, and transfer learning methods. The proposed meta-transfer learning achieves a relative character error rate (CER) reduction of 11.62% over meta-learning and a relative CER reduction of 10.86% over transfer learning in low-resource experiments. We used less than 15 minutes of data for each target language in near-zero resource language experiments. Our meta-transfer learning approach achieved an average CER of 25.25% less than meta-learning and transfer learning. These results clearly demonstrate that the proposed integration works well in ASR tasks in languages with very limited data resources.
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2024.3486711