Multitask Learning for Adaptive Quality Estimation of Automatically Transcribed Utterances

@inproceedings{Souza2015MultitaskLF,
  title={Multitask Learning for Adaptive Quality Estimation of Automatically Transcribed Utterances},
  author={Jos{\'e} Guilherme Camargo de Souza and Hamed Zamani and Matteo Negri and Marco Turchi and Daniele Falavigna},
  booktitle={HLT-NAACL},
  year={2015}
}
We investigate the problem of predicting the quality of automatic speech recognition (ASR) output under the following rigid constraints: i) reference transcriptions are not available, ii) confidence information about the system that produced the transcriptions is not accessible, and iii) training and test data come from multiple domains. To cope with these constraints (typical of the constantly increasing amount of automatic transcriptions that can be found on the Web), we propose a domain… CONTINUE READING

Citations

Publications citing this paper.

References

Publications referenced by this paper.
SHOWING 1-10 OF 37 REFERENCES

Confidence Estimation for Automatic Speech Recognition Hypotheses

  • Mathew Stephen Seigel
  • Ph.D. thesis,
  • 2013
1 Excerpt

Report on the 10th IWSLT Evaluation Campaign

  • Jan Niehues, Sebastian Stuker, Luisa Bentivogli, Marcello Federico
  • In Proceedings of the International Workshop on…
  • 2013

Similar Papers

Loading similar papers…