Coarse-To-Fine And Cross-Lingual ASR Transfer
This work addresses the challenge of ASR for low-resource languages like Czech by enabling efficient cross-lingual transfer, though it is incremental as it builds on existing transfer learning techniques.
The paper tackled the problem of training end-to-end neural automatic speech recognition (ASR) systems for low-resource languages by proposing a transfer learning method from English to Czech using an intermediate alphabet without accents, which reduced training time and word error rate (WER).
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).