Abstract
Canada is an officially bilingual country, but the only legal requirement is for federal services to be offered in both official languages. Therefore, services provided by provincial and municipal governments are typically offered only in the language of the majority, with cost being cited as the main obstacle to providing translation. This paper presents a recipient evaluation designed to determine whether machine translation could be used as a cost-effective means of increasing translation services in Canadian official language minority communities. The results show that not all communities have the same needs, and that raw or rapidly post-edited MT output is more suitable for information assimilation, while maximally post-edited MT output is a minimum requirement when translation is intended as a means of cultural preservation and promotion. The survey also suggests that average recipients are more receptive to MT than are language professionals.
Highlights
Evaluation of machine translation (MT) can and must take many forms depending both on the goal of the evaluation and on which stakeholders are involved (White, 2003)
The OCOL recommends that members of the official language minority communities (OLMCs) should be consulted to determine their needs and their level of satisfaction with proposed solutions. With these points in mind, we set out to conduct an investigation into whether MT can provide a faster and more cost-effective means of translation that would in turn enable provincial and municipal governments and agencies to offer a wider range of translation services to OLMCs
Once the various translations had been generated, the step was to consult members of each OLMC to try to identify the reasons why they wanted the texts to be made available in the other official language, and to determine which of the four translated versions presented to them (MT, rapidly post-edited (RPE), maximally post-edited (MPE) or human translation (HT)) best met their needs
Summary
Evaluation of machine translation (MT) can and must take many forms depending both on the goal of the evaluation and on which stakeholders are involved (White, 2003). The costs of providing translation services to OLMCs are considered, followed by description of an experiment that investigates whether MT could be used as a means of partially meeting the translation needs of these OLMCs. The experiment takes the form of parallel recipient evaluations conducted in the Saskatchewan and West Quebec OLMCs, and the results of these evaluations are discussed and compared. 16-17) in the Report on the Government of Canada’s Consultations on Linguistic Duality and Official Languages Given this situation, we set out to conduct a preliminary investigation into whether MT could offer at least a partial solution for addressing the unmet translation needs of Canadians living in OLMCs. A recipient evaluation is highly appropriate in this case because in order for use of MT to be considered viable, the intended recipients of the target texts must accept the output produced by MT systems. By gaining deeper understanding of some of the challenges facing these OLMCs, we can better interpret their reactions to MT output
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
