arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2006.05754cs.CLcs.AIcs.SDeess.AS
收藏

性别危险?基于MuST-SHE语料库的语音翻译技术评价

Gender in Danger? Evaluating Speech Translation Technology on the MuST-SHE Corpus

Fondazione Bruno Kessler(布鲁诺·凯斯勒基金会);University of Trento(特伦托大学)

Luisa Bentivogli, Beatrice Savoldi, Matteo Negri, Mattia Antonino Di Gangi, Roldano Cattoni, Marco Turchi

摘要:

从没有生产性语法性别的语言(如英语)翻译成性别标记的语言是机器的一个众所周知的困难。这种困难也是由于这样一个事实,即建立模型的训练数据通常反映了自然语言的不对称性,包括性别偏见。机器翻译只提供文本数据,因此在本质上受到输入句子并不总是包含有关所指人类实体性别身份的线索的事实的限制。但是,如果输入是音频信号,语音翻译会发生什么?音频能否提供更多信息以减少性别偏见?我们提出了第一次彻底调查的性别偏见在语音翻译,有助于:i)发布一个基准有用的未来的研究,和ii)不同的技术(级联和端到端)的比较两种语言方向(英语-意大利语/法语)。

英文摘要:

Translating from languages without productive grammatical gender like English into gender-marked languages is a well-known difficulty for machines. This difficulty is also due to the fact that the training data on which models are built typically reflect the asymmetries of natural languages, gender bias included. Exclusively fed with textual data, machine translation is intrinsically constrained by the fact that the input sentence does not always contain clues about the gender identity of the referred human entities. But what happens with speech translation, where the input is an audio signal? Can audio provide additional information to reduce gender bias? We present the first thorough investigation of gender bias in speech translation, contributing with: i) the release of a benchmark useful for future studies, and ii) the comparison of different technologies (cascade and end-to-end) on two language directions (English-Italian/French).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑