Meta launched a new language technology program with UNESCO. Within the scope of the program, speech records are collected by writing texts and translation data and it is aimed to develop artificial intelligence models. Meta announced that the information obtained will be used in the beautification of speech recognition and translation systems in different languages.

Meta launched a new attempt under the name of “Language Technology Cooperation Program”. Within the scope of the program, at least 10 hours of speech recording, written texts and business partners can provide translation sentences in different languages. The participants will work with Meta’s artificial intelligence groups and contribute to the integration of this information into speech recognition and translation models.

Meta will make its artificial intelligence open source

Meta announced that these models will be shared as open source when completed. In the midst of the participants of the program, Canada’s Nunavut government also includes. Some residents in the region speak of Inuit Languages ​​known as Inuktut. Meta states that it is heavier on languages ​​that are not represented in particular, and aims to work with UNESCO to develop artificial intelligence -based analyzes for more languages.

In addition to the program, Meta has announced an open source machine translation tool. Prepared by linguists and supporting seven languages, this vehicle was opened to access to developers. Users will be able to contribute to the vehicle via the artificial intelligence platform Huging Face. On the other hand, Meta said that he has been experiencing shortcomings in the past, while accepting that the spread of false information in language except English, he continued his efforts to beautify artificial intelligence -based moderation systems.