Translation services are useful for a variety of industries and organisations, of which international organisations play a major role. International Organisations require translating documents into multiple official languages, with high accuracy, within a limited time frame. Post-editing of Machine Translation (MT) is known to allow translating large volumes of translations while saving costs and time. Workflows in the translation industry have experienced a significant transformation and it is in this new context that speech technology is likely to contribute to further innovation. Preliminary studies show that provided that the ASR and MT output are of high quality and that the translators are competent with using software (Computer Assisted Translation (CAT) tools, MT suggestions and ASR toolkits such as Dragon), speech based post-editing can be a promising approach which can result in performance gains in the translation workflow.

In this paper we explore how different post-editing methods using speech and keyboard will perform in two different translation interfaces. Recent research encourages studies on multi-modal post-editing CAT environments combining different input possibilities including pen, touch and speech. However in international organisations, current CAT interfaces still heavily focus on traditional mouse and keyboard input to support PE operations. This opens up the question on whether using different modalities in different translation interfaces can be integrated to the translation process workflow in international organisations, which is an unexplored area of research so far.