Gender aware spoken language translation applied to English-Arabic

By Mostafa Elaraby, Ahmed Y. Tawfik, Mahmoud Khaled, Hany Hassan, A. Osama. 2018

47 citations recorded by the source corpus.

Spoken Language Translation (SLT) is becoming more widely used and becoming a communication tool that helps in crossing language barriers. One of the challenges of SLT is the translation from a language without gender agreement to a language with gender agreement such as English to Arabic. In this paper, we introduce an approach to tackle such limitation by enabling a Neural Machine Translation system to produce gender-aware unbiased translation. We show that NMT system can model the speaker/listener gender information to produce gender-aware translation that reduces the bias effect resulting from having training data dominated by particular gender forms. We propose a method to generate data used in adapting a NMT system to produce gender-aware and unbiased translation. The proposed approach can achieve significant improvement of the translation quality by 2 BLEU points.

Subfields: Artificial Intelligence, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

Citations in the corpus

Societal Biases in Language Generation: Progress and Challenges (2021, methodology): “…many use BLEU and its variants to evaluate correct gender inﬂections and translations (Moryossef et al., 2019; Escud´e Font and Costa-juss`a, 2019; Elaraby et al., 2018; Habash et al., 2019; Alhafni et al., 2020) or accuracy on WinoMT (Saunders and Byrne, 2020; Saunders et al., 2020; Kocmi et…”
Automatic Gender Identification and Reinflection in Arabic (2019, methodology): “Similarly, Elaraby et al. (2018) marked speaker and listener gender as meta-data input on the source sentence in an English-to-Arabic NMT system.”
A Survey on Gender Bias in Natural Language Processing (2021, result): “…et al. 2019; Vanmassenhove et al. 2018] cannot be ascribed with certainty to a better control of gender features and following previous research [Elaraby et al. 2018; Vanmassenhove et al. 2018] underlie the importance of applying gender-swapping in BLEU-based evaluations focused on gender…”
Mitigating Gender Bias in Machine Translation with Target Gender Annotations (2020, background): “Thus our approach is also simpler in terms of training data preparation compared to the previous work (Elaraby et al., 2018; Vanmassenhove et al., 2018).”
Gender Bias in Machine Translation (2021, background): “However, as in the case of (Vanmassenhove et al., 2018; Elaraby et al., 2018), it requires metadata about speakers and listeners.”
Improving Gender Translation Accuracy with Filtered Self-Training (2021, unknown): “biguous input or propose methods to control for gender or to augment data with gender information (Elaraby et al., 2018; Moryossef et al., 2019; Prates et al., 2018; Saunders et al., 2020; Stafanovičs et al., 2020; Vanmassenhove et al., 2018).”
Transfer Learning in Multilingual Neural Machine Translation with Dynamic Vocabulary (2018, methodology): “Inspired by [24], who automatically annotated the source side with artiﬁcial ﬂags to manage the politeness level of the output, other works focused on controlling the grammatical voice [25], the text domain [26, 27], and enforcing gender agreement [28].”