eduzhai > Applied Sciences > Engineering >

Text-Conditioned Transformer for Automatic Pronunciation Error Detection

  • king
  • (0) Download
  • 20210507
  • Save

... pages left unread,continue reading

Document pages: 13 pages

Abstract: Automatic pronunciation error detection (APED) plays an important role in thedomain of language learning. As for the previous ASR-based APED methods, thedecoded results need to be aligned with the target text so that the errors canbe found out. However, since the decoding process and the alignment process areindependent, the prior knowledge about the target text is not fully utilized.In this paper, we propose to use the target text as an extra condition for theTransformer backbone to handle the APED task. The proposed method can outputthe error states with consideration of the relationship between the inputspeech and the target text in a fully end-to-end fashion.Meanwhile, as theprior target text is used as a condition for the decoder input, the Transformerworks in a feed-forward manner instead of autoregressive in the inferencestage, which can significantly boost the speed in the actual deployment. We setthe ASR-based Transformer as the baseline APED model and conduct severalexperiments on the L2-Arctic dataset. The results demonstrate that our approachcan obtain 8.4 relative improvement on the $F 1$ score metric.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×