Speech recognition has long been used to support dictation in translation, while machine translation (MT) post-editing (PE) has become an established part of many translators' workflows. However, little attention has been paid to the use of speech recognition during PE itself. To better understand translators' views on this possibility, we conducted an online survey exploring current practices, perceived benefits and barriers, and attitudes towards speech-assisted machine translation post-editing (MTPE). After quality control, responses from 52 participants were analysed using descriptive statistics, together with an analysis of the open-ended responses.
Although a promising 53.8% of respondents had used speech recognition for purposes outside translation, only 3.8% reported using it for MTPE. Respondents generally associated speech recognition with practical advantages such as speed and improved ergonomics, with 69.5% identifying ergonomic benefits as a motivation. At the same time, only 36.5% expressed openness towards speech-assisted PE, while 72.8% considered continuing to work by typing to be easier. Comments provided in the open-ended questions suggested that speech recognition was seen as particularly useful for specific situations—such as repetitive revisions, short segments, or voice commands within CAT tools—rather than as a complete replacement for keyboard-based PE.
Taken together, these findings show that speech recognition is seen as a useful complement to existing PE workflows, particularly where it can improve ergonomics or support specific editing tasks, rather than replace keyboard-based interaction. They also highlight the need to explore new forms of speech-based PE, including using spoken natural-language commands to guide large language models during the editing process. These findings informed the design of COPECO-Speech, a multimodal research and teaching platform that integrates typing, speech-based navigation, and LLM-mediated spoken editing instructions. The platform provides a foundation for systematically investigating how speech interaction can be integrated into future MTPE workflows while also supporting translator education.
Jeevanthi Liyanapathirana is a PhD student at the Faculty of Translation and Interpreting, University of Geneva, where her research question lies on incorporating speech technologies for translation and post editing purposes. She has been a fellow in translation technology as well as a translation technologist in the World Intellectual Property Organization, Geneva and is currently working as an IT Solutions Analyst at the World Trade Organization, Geneva, Switzerland. She holds a Masters of Philosophy in Computational Linguistics from the University of Cambridge, UK (MPhil in Computer Speech, Text and Internet Technology) and a Bachelor of Science (Computer Science Special Degree) from the University of Colombo, Sri Lanka.
Pierrette Bouillon has been Professor at the Faculty of Translation and Interpreting(FTI), University of Geneva since 2007. She is currently Director of the Department of Translation Technology (referred to by its French acronym TIM) and Dean of the FTI. She has numerous publications in computational linguistics and natural language processing, particularly within speech-to-speech machine translation, accessibility and pre-editing/post-editing.
