Efficient Utilization of Large Pre-Trained Models for Low Resource ASR
This work addresses the problem of adapting large pre-trained models for low-resource ASR in specific domains like medical telephony, representing an incremental improvement with practical applications.
The paper tackled low-resource automatic speech recognition for conversational telephony speech in Vietnamese and German by efficiently utilizing large pre-trained models, achieving a 22% relative improvement over baselines with pretraining techniques and additional gains of up to 29% through architectural refinements.
Unsupervised representation learning has recently helped automatic speech recognition (ASR) to tackle tasks with limited labeled data. Following this, hardware limitations and applications give rise to the question how to take advantage of large pre-trained models efficiently and reduce their complexity. In this work, we study a challenging low resource conversational telephony speech corpus from the medical domain in Vietnamese and German. We show the benefits of using unsupervised techniques beyond simple fine-tuning of large pre-trained models, discuss how to adapt them to a practical telephony task including bandwidth transfer and investigate different data conditions for pre-training and fine-tuning. We outperform the project baselines by 22% relative using pretraining techniques. Further gains of 29% can be achieved by refinements of architecture and training and 6% by adding 0.8 h of in-domain adaptation data.