Microsoft India announced the addition of three new Indian languages Konkani, Maithili, and Sindhi, in Microsoft Translator. Along with this, Microsoft Translator will also support Sinhala, the official language of Sri Lanka. With this latest release, Microsoft is further democratizing access to information in native languages for India. Microsoft Translator now supports 16 Indian languages: including Assamese, Bengali, Gujarati, Hindi, Kannada, Konkani, Maithili, Malayalam, Marathi, Nepali, Odia, Punjabi, Sindhi, Tamil, Telugu, and Urdu.
Microsoft Translator will now allow over 95% of Indians to access information and work in their native or preferred languages, making computing language-agnostic and more inclusive in India. It can help general users interpret real-time conversations, menus, street signs, websites, and documents. Companies can leverage it to globalize their business and strengthen customer outreach. Microsoft Translator can be used across Windows, iOS, Android, and the web. Microsoft also continues to push the boundaries on the quality of translations across Indian languages.
The function is available on the Microsoft Translator app, edge browser, Office 365, Bing Translator, and through the Azure Cognitive Services Translator API for businesses and developers. Users can translate Konkani, Maithili, Sindhi, and Sinhala text, supported in more than 125 languages, for their apps, websites, workflows, and tools with Azure Cognitive Services Translator. Businesses can also avail multi-language support such as translation for e-content, e-commerce product catalogues, product documentation, and internal communication, among others.
Konkani is spoken by over 2 million people in India, primarily in the states of Goa, Maharashtra, and Karnataka. It is also spoken by a significant number of people in other parts of India, such as Kerala and Gujarat. Maithili is spoken by over 75 million people in India and Nepal. It is the second most widely spoken language in the Indian state of Bihar and is also spoken in the neighboring states of Jharkhand and West Bengal. The development of Maithili was championed by Dr. Girish Nath Jha from the Jawaharlal Nehru University (JNU) New Delhi, who provided critical test sets and initial training data.
Sindhi is spoken by over 20 million people in India, and several other countries in the subcontinent while Sinhala is spoken by over 16 million people in Sri Lanka, as well as in other countries such as Malaysia and Singapore.
Rajiv Kumar, Managing Director – Microsoft India, said, “At Microsoft we are committed to empower every individual and organization on the planet and this commitment is reflected in our attempt to broaden the reach of technology and inclusiveness through language as a medium. We are excited to announce that we are broadening our language capabilities to include Maithili, Konkani, Sindhi, and Sinhala. We celebrate and support India’s diversity of language and culture with the most advanced AI to enable India’s growth, by making access to technology pervasive. “
Bringing Deep Neural Networks to language translation
Microsoft has been using Deep Neural Networks to develop language models for translating and transliterating complex Indian languages. Deep Neural Networks are also sensitive to minor linguistic nuances such as gender (feminine, masculine, neutral), politeness level (slang, casual, written, formal), and word type (verb, noun, adjective). Microsoft continuously improves the translation models in line with tech advancement, usage, and releases newer, improved versions to all its users in a transparent manner.