Can Real-Time Machine Translation Overcome Language Barriers in Distributed Requirements Engineering?

Abstract

In global software projects work takes place over long distances, meaning that communication will often involve distant cultures with different languages and communication styles that, in turn, exacerbate communication problems. However, being aware of cultural distance is not sufficient to overcome many of the barriers that language differences bring in the way of global project success. In this paper, we investigate the adoption of machine translation (MT) services in synchronous text-based chat in order to overcome any language barrier existing among groups of stakeholders who are remotely negotiating software requirements. We report our findings from a simulated study that compares the efficiency and the effectiveness of two MT services, Google Translate and apertium-service, in translating the messages exchanged during four distributed requirements engineering workshops. The results show that (a) Google Translate produces significantly more adequate translations than Apertium from English to Italian; (b) both services can be used in text-based chat without disrupting real-time interaction

A Controlled Experiment on the Effects of Machine Translation in Multilingual Requirements Meetings

“…), based on the data collected so far, we have found evidence that the use of MT is accepted with favor by participants and is not disruptive of the conversation flow, even during the execution of complex group tasks, such as distributed requirements meetings. Such finding is interesting because, as already shown by our previous study [6], state-of-the-art MT services are still far from 100% accuracy. Thus, the point is whether meeting participants always require exact word translations as long as it is still understood to a large extent.…”

“…As of this writing, Google Translate supports the translation between any two pairs of over 50 languages, although not all at the same quality level. In our previous work [6], according to a set of human raters, Google Translate was found to produce better (i.e. more accurate) automatic translation than the rule-based Apertium 3 service.…”

“…In our previous work [6] we run a simulated study to evaluate the feasibility of adopting an automatic, crosslanguage translation to communication-intensive activities, such as distributed requirements engineering. Although our work proved that state-of-the-art machine translation services could be embedded into synchronous text-based chat with a negligible extra time, being only a simulation, the study did not allow us to hypothesize whether the quality of machine translation services would be good enough to allow participants to complete a complex group task while communicating with their own native language.…”

An empirical simulation-based study of real-time speech translation for multilingual global project teams

“…We first run a simulated study, which proved that state-ofthe-art machine translation services, such as Google Translate, could be embedded into synchronous text-based chat with a negligible extra time [5]. Then, we conducted a controlled experiment [6] and a replication [7] to investigate whether realtime machine translation could be successfully used instead of English in distributed multilingual requirements meetings, with non-native speakers with different level of proficiency.…”

Multilingual Chat through Machine Translation: A Case of English-Russian

“…But machine translation worked better with some languages such as Italian, Serbian and Russian, than with some others like Filipino, Japanese and Hindi. Calefato, Lanubile and Minervini (2010) investigated "the adoption of machine translation (MT) services in a synchronous text-based chat in order to prevail over language barriers when stakeholders are remotely negotiating sotware requirements." They used two real-time MT services: Google Translate and Apertium.…”