The impact of using text-to-speech (TTS) in post-editing machine translation (PEMT) workflows on translators’ cognitive effort, productivity, quality, and perceptions
This eye-tracking experiment investigates the impact of text-to-speech (TTS) alongside translation and post-editing machine translation (PEMT) experience on translators’ cognitive effort, productivity, quality, and perceptions during PEMT from English into German. TTS substantially lowered PEMT speed and target text mean fixation duration, and increased source and target text fixation counts and dwell time. TTS had no substantial effect on edit distance or target text quality. Experience had no substantial effect. 14 out of 21 participants preferred post-editing with TTS.
Authors
- Jan Martin Brockmann (ORCID: https://orcid.org/0000-0002-7647-4515)
- Dragoș Ciobanu (ORCID: https://orcid.org/0000-0002-6475-080X)
- Alina Secară (ORCID: https://orcid.org/0000-0001-6281-5035)
- Miguel Rios (ORCID: https://orcid.org/0009-0000-6823-9032)
- Raluca-Maria Chereji (ORCID: https://orcid.org/0000-0001-6112-0856)
- Claudia Wiesinger
Institutions
- University of Vienna (AT)
Publication Details
- Journal
- Tradumàtica tecnologies de la traducció
- Published
- 2024-12-31
- DOI
- https://doi.org/10.5565/rev/tradumatica.394
- Primary Topic
- Natural Language Processing Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00