Evaluating Logit-Based GOP Scores for Mispronunciation Detection

This repository contains resources for the paper:

"Evaluating Logit-Based GOP Scores for Mispronunciation Detection"
by Aditya Kamlesh Parikh, Cristian Tejedor-Garcia, Catia Cucchiarini, and Helmer Strik

📣 Accepted at Interspeech 2025!

Citation

If you use this work, please cite it as:

📄 LaTeX (BibTeX)

@inproceedings{parikh25b_interspeech,
  title     = {{Evaluating Logit-Based GOP Scores for Mispronunciation Detection}},
  author    = {Aditya Kamlesh Parikh and Cristian Tejedor-Garcia and Catia Cucchiarini and Helmer Strik},
  year      = {2025},
  booktitle = {{Interspeech 2025}},
  pages     = {2405--2409},
  doi       = {10.21437/Interspeech.2025-1012},
  issn      = {2958-1796},
}

📚 APA

Parikh, A.K., Tejedor-Garcia, C., Cucchiarini, C., Strik, H. (2025) Evaluating Logit-Based GOP Scores for Mispronunciation Detection. Proc. Interspeech 2025, 2405-2409, doi: 10.21437/Interspeech.2025-1012

Acknowledgements

Part of the project Responsible AI for Voice Diagnostics (RAIVD) with file number NGF.1607.22.013 of the research program NGF AiNed Fellowship Grants, which is financed by the Dutch Research Council (NWO).