Due to the lack of information required for automated training systems, processing minority language content has been considered difficult for many translation and retrieval technologies until recently.
Raytheon BBN’s multilingual information retrieval technology enables English speakers to search text and analyze segments of speech in a foreign language, as well as obtain relevant summaries translated from a low-resource foreign language to the English language. ‘English. The application facilitates the retrieval of information in text or speech form, as well as the translation of words and sentences, and helps in understanding the context and meaning of material without the need to learn a new language.
According to stripes.com, the company’s breakthrough solution allows users to search text and spoken snippets in multiple languages using English search phrases. The tool was produced as part of the IAPRA MATERIAL initiative and took four years to build.
Raytheon’s new technology allows users to obtain documents and records in other languages for use in English queries, as well as providing expert translation of relevant terms. The user can grasp context and meaning without having to learn the language in this way. Kazakh, Pashto, Swahili, Somali, and Tagalog were used to demonstrate system performance, even in languages with limited machine learning resources, such as