Google invests a lot in its voice neural machine learning and now announced that it added support for 30 new languages. It has a thousand-year old Georgian, two of the biggest African languages and plethora of Indian languages.
To incorporate all languages, Google worked with natives and collected speech samples. Using the specific sounds and words, the machines trained themselves to understand and improve accuracy when exposed to more examples over time.
The new languages are available in Cloud Speech API. Soon they will start appearing across Google apps like Google Translate and Gboard keyboard. To use them, the Voice Search feature must be enabled.
Here’s the full list of newly supported languages:
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).