Abstract
Disfluencies that appear in the transcriptions from automatic speech recognition systems tend to impair the performance of downstream NLP tasks. Disfluency correction models can help alleviate this problem. However, the unavailability of labeled data in low-resource languages impairs progress. We propose using a pretrained multilingual model, finetuned only on English disfluencies, for zero-shot disfluency detection in Indian languages. We present a detailed pipeline to synthetically generate disfluent text and create evaluation datasets for four Indian languages: Bengali, Hindi, Malayalam, and Marathi. Even in the zero-shot setting, we obtain F1 scores of 75 and higher on five disfluency types across all four languages. We also show the utility of synthetically generated disfluencies by evaluating on real disfluent text in Bengali, Hindi, and Marathi. Finetuning the multilingual model on additional synthetic Hindi disfluent text nearly doubles the number of exact matches and yields a 20-point boost in F1 scores when evaluated on real Hindi disfluent text, compared to training with only English disfluent text.- Anthology ID:
- 2022.coling-1.392
- Volume:
- Proceedings of the 29th International Conference on Computational Linguistics
- Month:
- October
- Year:
- 2022
- Address:
- Gyeongju, Republic of Korea
- Venue:
- COLING
- SIG:
- Publisher:
- International Committee on Computational Linguistics
- Note:
- Pages:
- 4442–4454
- Language:
- URL:
- https://aclanthology.org/2022.coling-1.392
- DOI:
- Cite (ACL):
- Rohit Kundu, Preethi Jyothi, and Pushpak Bhattacharyya. 2022. Zero-shot Disfluency Detection for Indian Languages. In Proceedings of the 29th International Conference on Computational Linguistics, pages 4442–4454, Gyeongju, Republic of Korea. International Committee on Computational Linguistics.
- Cite (Informal):
- Zero-shot Disfluency Detection for Indian Languages (Kundu et al., COLING 2022)
- PDF:
- https://preview.aclanthology.org/auto-file-uploads/2022.coling-1.392.pdf
- Data
- PMIndia