Integrating Language Models into Direct Speech Translation: An Inference-Time Solution to Control Gender Inflection
Dennis Fucci, Marco Gaido, Sara Papi, Mauro Cettolo, Matteo Negri, Luisa Bentivogli
Abstract
When translating words referring to the speaker, speech translation (ST) systems should not resort to default masculine generics nor rely on potentially misleading vocal traits. Rather, they should assign gender according to the speakers’ preference. The existing solutions to do so, though effective, are hardly feasible in practice as they involve dedicated model re-training on gender-labeled ST data. To overcome these limitations, we propose the first inference-time solution to control speaker-related gender inflections in ST. Our approach partially replaces the (biased) internal language model (LM) implicitly learned by the ST decoder with gender-specific external LMs. Experiments on en→es/fr/it show that our solution outperforms the base models and the best training-time mitigation strategy by up to 31.0 and 1.6 points in gender accuracy, respectively, for feminine forms. The gains are even larger (up to 32.0 and 3.4) in the challenging condition where speakers’ vocal traits conflict with their gender.- Anthology ID:
- 2023.emnlp-main.705
- Volume:
- Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Houda Bouamor, Juan Pino, Kalika Bali
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 11505–11517
- Language:
- URL:
- https://aclanthology.org/2023.emnlp-main.705
- DOI:
- 10.18653/v1/2023.emnlp-main.705
- Cite (ACL):
- Dennis Fucci, Marco Gaido, Sara Papi, Mauro Cettolo, Matteo Negri, and Luisa Bentivogli. 2023. Integrating Language Models into Direct Speech Translation: An Inference-Time Solution to Control Gender Inflection. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 11505–11517, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- Integrating Language Models into Direct Speech Translation: An Inference-Time Solution to Control Gender Inflection (Fucci et al., EMNLP 2023)
- PDF:
- https://preview.aclanthology.org/emnlp-22-attachments/2023.emnlp-main.705.pdf