Coarse-To-Fine And Cross-Lingual ASR Transfer
arXiv:2109.00916
Abstract
End-to-end neural automatic speech recognition systems achieved recently state-of-the-art results, but they require large datasets and extensive computing resources. Transfer learning has been proposed to overcome these difficulties even across languages, e.g., German ASR trained from an English model. We experiment with much less related languages, reusing an English model for Czech ASR. To simplify the transfer, we propose to use an intermediate alphabet, Czech without accents, and document that it is a highly effective strategy. The technique is also useful on Czech data alone, in the style of coarse-to-fine training. We achieve substantial eductions in training time as well as word error rate (WER).
arXiv commentAccepted to ITAT WAFNL
References in corpus (5)
- SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
- Improved Regularization of Convolutional Neural Networks with Cutout
- A Comparative Study on Transformer vs RNN in Speech Applications
- Stochastic Gradient Methods with Layer-wise Adaptive Moments for Training of Deep Networks
- Cross-Language Transfer Learning, Continuous Learning, and Domain Adaptation for End-to-End Automatic Speech Recognition