As of today, Google has added dozens of new languages to its speech recognition software. Most of these newly added languages are from regions inside Africa and India as both countries are one of the biggest emerging markets. This will let more and more people surf the web through voice typing.
According to Google, the latest update will add 30 more languages to the speech recognition which will make a total of 119 supported languages. The new update will feature 8 more Indian languages along with Africa’s two largest languages; Swahili and Amharic.
However, these new languages in speech recognition will be available only in the Gboard voice typing and not for Google Now in order to make the Gboard app all the more famous and the most used keyboard app out there. Moreover, there is good news for those who speak the U.S. English as they can use their voice to use emojis instead of searching for them by just saying the emoji name such as, “winky face emoji” and the emoji will be added to the text field. These new languages are also available in the Cloud Speech API. The API already supports 89 languages and is used in a number of third-party voice recognition enabled apps.
Also Read: Google Gboard’s latest update: Roman Urdu to Urdu script conversion
India and Africa; emerging markets
Although regions like India and Africa are neglected when it comes to giving them access to new technology, tech companies are now aiming to establish platforms in these countries by providing them with the latest technologies. In India, there is a lot to gain as the country has more than 420 million mobile internet users. This makes India the largest market for tech companies to invest and progress in. This is the main reason why Google has added 8 languages from India to its speech recognition as this move will have a significant impact on the company’s progress.
Google’s support for new languages came in as the company is seeing a new kind of internet users coming in the future. As it is expected that there will be 1 billion new internet users, they will use all the latest tech the internet has to offer. In such a time period, adding support for more and more languages will prove to be all the more beneficial for the company.
Quick Read: With UrduScript, You can now learn JavaScript in Roman Urdu
How it works
Google has developed these language packs by combining human labor with its machine learning technologies. The workers are asked to read common phrases of their native languages and that’s how Google collects speech samples. These samples are then used to train machines understand the languages by listening to the sound and accent of the phrases. From time to time, new languages are added and the older ones are updated thereby increasing the efficiency of the machines.
Here is a list of the newly added languages by Google:
- 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)
Source: TechCrunch