Microsoft has added Hmong to the list of languages supported by Bing Translator. This was achieved using new features of Microsoft Translator to train a new translation engine using parallel documents.
Hmong is spoken by over 6.5 million people mainly in Southeast Asia and China but also by communities in the United States, Australia, and France. The dialect supported by Bing Translator is Hmong Daw.
Hmong has a very long oral tradition, but the writing system was developed just 60 years ago. The Hmong community, which was instrumental in getting this language added to Bing, is concerned with the long-term preservation of the language and sees an important role for online machine translation in this context.
Members of the community used new tools currently in beta to train the new translation engine.
Microsoft Translator ... can learn how to translate from a set of parallel documents (same document in two languages), dictionaries and texts in the language to translate to (Hmong in this case).
In addition to teaching the engine a new language, they also involved members of the community, partners and collaborators to create and review improved versions of the automated translation system, and collect qualitative feedback about each “trained” system. Deploying a system that reaches a certain level of quality allows seamless use with the standard Microsoft Translator APIs, and many scenarios powered by the API, like the web translation widget. Feedback that is generated through these scenarios can be utilized again in the training process – creating a virtuous loop for improving the translation quality.
(click to enlarge)
Microsoft added this latest language on February 21st in celebration of International Mother Language Day.
According to its page on the United Nations website , International Mother Language Day has been observed in February every year since 2000 to promote linguistic and cultural diversity and multilingualism. The date represents the day in 1952 when students demonstrating for recognition of their language, Bangla, as one of the two national languages of the then Pakistan, were shot and killed by police in Dhaka, the capital of what is now Bangladesh.
There are now 38 languages available using Bing Translator, which leaves a lot of scope for more to be added.
You can sign up for the Microsoft Translator API using Windows Azure Marketplace and can try out the functionality of the Translator API, without having to write any code, on the Marketplace itself.
The only real problem with the previous release of Google's TensorFlow was that it would only work on one computer. Training neural networks is computationally intensive so the good news is that the l [ ... ]