ISCA Archive Interspeech 2005
ISCA Archive Interspeech 2005

Open vocabulary speech recognition with flat hybrid models

Maximilian Bisani, Hermann Ney

Today's speech recognition systems are able to recognize arbitrary sentences over a large but finite vocabulary. However, many important speech recognition tasks feature an open, constantly changing vocabulary. (E.g. broadcast news transcription, translation of political debates, etc. Ideally, a system designed for such open vocabulary tasks would be able to recognize arbitrary, even previously unseen words. To some extent this can be achieved by using sub-lexical language models. We demonstrate that, by using a simple flat hybrid model, we can significantly improve a well-optimized state-of-the-art speech recognition system over a wide range of out-of-vocabulary rates.

doi: 10.21437/Interspeech.2005-11

Cite as: Bisani, M., Ney, H. (2005) Open vocabulary speech recognition with flat hybrid models. Proc. Interspeech 2005, 725-728, doi: 10.21437/Interspeech.2005-11

  author={Maximilian Bisani and Hermann Ney},
  title={{Open vocabulary speech recognition with flat hybrid models}},
  booktitle={Proc. Interspeech 2005},