Recent advances in NMT have improved the fluency of the output in certain language pairs and domains to such an extent that translators may be lured into a "fluency trap" during machine translation post-editing (MTPE). This can cause post-editors to miss significant errors. This session will describe research into the use of speech-enabled post-editing modalities, specifically text-to-speech technology during MTPE. We will summarize the foundational research and look at the impacts on post-editors' performance. Additionally, we will discuss preliminary results of an industry-funded research project that uses eye-tracking, alongside other methods, to measure the practicality and cognitive impact of this type of MTPE approach.