Google’s speech recognition can now understand 30 more languages.
Google’s voice search is becoming more inclusive by picking up support for 30 new languages, including eight Indian languages. In all, over 1 billion people can now use voice dictation on Google’s services, including the Google app and Gboard. Notable additions include popular Indian dialects like Gujarati, Bengali, Kannada, Marathi, Tamil, and Telugu, along with two of Africa’s most-spoken languages: Swahili and Amharic.
Here’s a list of all the new additions:
- Amharic (Ethiopia)
- Armenian (Armenia)
- Azerbaijani (Azerbaijani)
- Bengali (Bangladesh, India)
- English (Ghana, Kenya, Nigeria, Tanzania)
- Georgian (Georgia)
- Gujarati (India)
- Javanese (Indonesia)
- Kannada (India)
- Khmer (Cambodian)
- Lao (Laos)
- Latvian (Latvia)
- Malayalam (India)
- Marathi (India)
- Nepali (Nepal)
- Sinhala (Sri Lanka)
- Sundanese (Indonesia)
- Swahili (Tanzania, Kenya)
- Tamil (India, Singapore, Sri Lanka, Malaysia)
- Telugu (India)
- Urdu (Pakistan, India)
To enable for these languages, Google worked with native speakers, collecting speech samples and asking them to read common phrases so it can train its machine learning models to pick up on intonation.
The server-side update is now live, and you can set your language in the Google app or on Gboard to get started. I was able to ask Google a question in Telugu and it understood the query perfectly. The new additions are now available in the Cloud Speech API as well, and will be added to other Google services — including Translate — shortly.
Finally, if you’re using English U.S. as your language, you now have the option to search for emoji with your voice. For instance, you can just search for “rolling on the floor laughing emoji” on Gboard and it’ll surface ?. Technology is a wonderful thing.
