Conference paper Open Access

On Using SpecAugment for End-to-End Speech Translation

Bahar, Parnia; Zeyer, Albert; Schlüter, Ralf; Ney, Hermann

Citation Style Language JSON Export

  "publisher": "Zenodo", 
  "DOI": "10.5281/zenodo.3525010", 
  "author": [
      "family": "Bahar, Parnia"
      "family": "Zeyer, Albert"
      "family": "Schl\u00fcter, Ralf"
      "family": "Ney, Hermann"
  "issued": {
    "date-parts": [
  "abstract": "<p>This work investigates a simple data augmentation technique, SpecAugment, for end-to-end speech translation. SpecAugment is a low-cost implementation method applied directly to the audio input features and it consists of masking blocks of frequency channels, and/or time steps. We apply SpecAugment on end-to-end speech translation tasks and achieve up to +2.2% BLEU&nbsp;on LibriSpeech Audiobooks En&rarr;Fr and +1.2% on IWSLT TED-talks En&rarr;De by alleviating overfitting to some extent. We also examine the effectiveness of the method in a variety of data scenarios and show that the method also leads to significant improvements in various data conditions irrespective of the amount of training data.</p>", 
  "title": "On Using SpecAugment for End-to-End Speech Translation", 
  "type": "paper-conference", 
  "id": "3525010"
All versions This version
Views 109107
Downloads 8383
Data volume 63.8 MB63.8 MB
Unique views 9593
Unique downloads 7171


Cite as