調查計算語言文檔雙語方法中的語言影響

• 2020 年 4 月 3 日
• 筆記

對於瀕危語言而言，數據收集活動必須能夠應對很多數據源自口傳而且生產副本費用高昂的挑戰。因此，為了確保錄音的可解釋性，至少要將這些錄音轉譯成使用廣泛的語言版本。本文中，我們對翻譯語言的選擇如何影響記錄後的工作以及可能的自動方法方面進行了研究，這些自動方法會影響產生的雙語語料庫。為了解翻譯語言選擇對這些工作和方法的影響，我們採用MaSS多語言語音語料庫（Boito等人，2020）創建了56個雙語對並將這些雙語對應用到了資源缺乏的無監管詞切分和詞切分任務中。研究結果中重點強調了翻譯語言的選擇對詞切分性能的影響而且利用不同的已對齊譯文會學到不同的辭彙。最後，本文提出了一種雙語詞切分的混合方法，這種方法將從非參數貝葉斯模型中摘錄的範圍提示（Goldwater等人，2009a）與Godard等人（2018）的注意詞切分網路模型組合在一起。研究結果表明，將這些提示整合到網路模型的輸入表示中能夠提高翻譯和對齊品質，尤其是非常複雜的語言對。

原文標題：Investigating Language Impact in Bilingual Approaches for Computational Language Documentation

For endangered languages, data collection campaigns have to accommodate the challenge that many of them are from oral tradition, and producing transcriptions is costly. Therefore, it is fundamental to translate them into a widely spoken language to ensure interpretability of the recordings. In this paper we investigate how the choice of translation language affects the posterior documentation work and potential automatic approaches which will work on top of the produced bilingual corpus. For answering this question, we use the MaSS multilingual speech corpus (Boito et al., 2020) for creating 56 bilingual pairs that we apply to the task of low-resource unsupervised word segmentation and alignment. Our results highlight that the choice of language for translation influences the word segmentation performance, and that different lexicons are learned by using different aligned translations. Lastly, this paper proposes a hybrid approach for bilingual word segmentation, combining boundary clues extracted from a non-parametric Bayesian model (Goldwater et al., 2009a) with the attentional word segmentation neural model from Godard et al. (2018). Our results suggest that incorporating these clues into the neural models' input representation increases their translation and alignment quality, specially for challenging language pairs.

原文作者：Marcely Zanon Boito, Aline Villavicencio, Laurent Besacier

原文鏈接：https://arxiv.org/abs/2003.13325