It's Not a Walk in the Park! Challenges of Idiom Translation in Speech-to-text Systems

• URL: http://arxiv.org/abs/2506.02995v1
• Date: Tue, 03 Jun 2025 15:29:52 GMT
• Title: It's Not a Walk in the Park! Challenges of Idiom Translation in Speech-to-text Systems
• Authors: Iuliia Zaitova, Badr M. Abdullah, Wei Xue, Dietrich Klakow, Bernd Möbius, Tania Avgustinova,
• Abstract summary: We evaluate translation as compared to conventional news translation in both text-to-text machine translation (MT) and speech-to-text translation (SLT) systems.<n>Our results reveal that SLT systems experience a pronounced performance drop on idiomatic data, often reverting to literal translations even in higher layers.
• Score: 26.39440238965029
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Idioms are defined as a group of words with a figurative meaning not deducible from their individual components. Although modern machine translation systems have made remarkable progress, translating idioms remains a major challenge, especially for speech-to-text systems, where research on this topic is notably sparse. In this paper, we systematically evaluate idiom translation as compared to conventional news translation in both text-to-text machine translation (MT) and speech-to-text translation (SLT) systems across two language pairs (German to English, Russian to English). We compare state-of-the-art end-to-end SLT systems (SeamlessM4T SLT-to-text, Whisper Large v3) with MT systems (SeamlessM4T SLT-to-text, No Language Left Behind), Large Language Models (DeepSeek, LLaMA) and cascaded alternatives. Our results reveal that SLT systems experience a pronounced performance drop on idiomatic data, often reverting to literal translations even in higher layers, whereas MT systems and Large Language Models demonstrate better handling of idioms. These findings underscore the need for idiom-specific strategies and improved internal representations in SLT architectures.

Related papers

• Trans-Zero: Self-Play Incentivizes Large Language Models for Multilingual Translation Without Parallel Data [64.4458540273004]
  We propose a self-play framework that leverages only monolingual data and the intrinsic multilingual knowledge of Large Language Models (LLMs)<n>Experiments demonstrate that this approach not only matches the performance of models trained on large-scale parallel data but also excels in non-English translation directions.
  arXiv  Detail & Related papers  (2025-04-20T16:20:30Z)
• Wav2Gloss: Generating Interlinear Glossed Text from Speech [78.64412090339044]
  We propose Wav2Gloss, a task in which four linguistic annotation components are extracted automatically from speech. We provide various baselines to lay the groundwork for future research on Interlinear Glossed Text generation from speech.
  arXiv  Detail & Related papers  (2024-03-19T21:45:29Z)
• Crossing the Threshold: Idiomatic Machine Translation through Retrieval Augmentation and Loss Weighting [66.02718577386426]
  We provide a simple characterization of idiomatic translation and related issues. We conduct a synthetic experiment revealing a tipping point at which transformer-based machine translation models correctly default to idiomatic translations. To improve translation of natural idioms, we introduce two straightforward yet effective techniques.
  arXiv  Detail & Related papers  (2023-10-10T23:47:25Z)
• Towards Effective Disambiguation for Machine Translation with Large Language Models [65.80775710657672]
  We study the capabilities of large language models to translate "ambiguous sentences" Experiments show that our methods can match or outperform state-of-the-art systems such as DeepL and NLLB in four out of five language directions.
  arXiv  Detail & Related papers  (2023-09-20T22:22:52Z)
• Do GPTs Produce Less Literal Translations? [20.095646048167612]
  Large Language Models (LLMs) have emerged as general-purpose language models capable of addressing many natural language generation or understanding tasks. We find that translations out of English (E-X) from GPTs tend to be less literal, while exhibiting similar or better scores on Machine Translation quality metrics.
  arXiv  Detail & Related papers  (2023-05-26T10:38:31Z)
• Discourse Centric Evaluation of Machine Translation with a Densely Annotated Parallel Corpus [82.07304301996562]
  This paper presents a new dataset with rich discourse annotations, built upon the large-scale parallel corpus BWB introduced in Jiang et al. We investigate the similarities and differences between the discourse structures of source and target languages. We discover that MT outputs differ fundamentally from human translations in terms of their latent discourse structures.
  arXiv  Detail & Related papers  (2023-05-18T17:36:41Z)
• A Semi-supervised Approach for a Better Translation of Sentiment in Dialectical Arabic UGT [2.6763498831034034]
  We introduce a semi-supervised approach that exploits both monolingual and parallel data for training an NMT system. We will show that our proposed system can significantly help with correcting sentiment errors detected in the online translation of dialectical Arabic UGT.
  arXiv  Detail & Related papers  (2022-10-21T11:55:55Z)
• Can Transformer be Too Compositional? Analysing Idiom Processing in Neural Machine Translation [55.52888815590317]
  Unlike literal expressions, idioms' meanings do not directly follow from their parts. NMT models are often unable to translate idioms accurately and over-generate compositional, literal translations. We investigate whether the non-compositionality of idioms is reflected in the mechanics of the dominant NMT model, Transformer.
  arXiv  Detail & Related papers  (2022-05-30T17:59:32Z)
• Why don't people use character-level machine translation? [69.53730499849023]
  Despite evidence that character-level systems are comparable with subword systems, they are virtually never used in competitive setups in machine translation competitions. Character-level MT systems show neither better domain robustness, nor better morphological generalization, despite being often so motivated.
  arXiv  Detail & Related papers  (2021-10-15T16:43:31Z)
• Can You Traducir This? Machine Translation for Code-Switched Input [0.0]
  Code-Switching (CSW) is a common phenomenon that occurs in multilingual geographic or social contexts. We focus here on Machine Translation (MT) of CSW texts, where we aim to simultaneously disentangle and translate the two mixed languages. Experiments show this training strategy yields MT systems that surpass multilingual systems for code-switched texts.
  arXiv  Detail & Related papers  (2021-05-11T08:06:30Z)
• Can Your Context-Aware MT System Pass the DiP Benchmark Tests? : Evaluation Benchmarks for Discourse Phenomena in Machine Translation [7.993547048820065]
  We introduce the first of their kind MT benchmark datasets that aim to track and hail improvements across four main discourse phenomena. Surprisingly, we find that existing context-aware models do not improve discourse-related translations consistently across languages and phenomena.
  arXiv  Detail & Related papers  (2020-04-30T07:15:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.