Google Translator begins offering the possibility of translation in 24 more languages as of this Wednesday (11), which, according to the company, are deprived of the technology. The highlight is Guarani, which is spoken in Paraguay, as well as indigenous populations in Brazil, Argentina, and Chile.
Along with Guarani, which is one of Mercosur’s official languages and is spoken by about 7 million people worldwide, it will also be added to the Quechua and Aymara translation platform, spoken by native peoples of Peru, Bolivia and Argentina. An estimated 10 million and 2 million, respectively, speak these two languages.
How technology learned languages
According to Isaac Caswell, a software engineer and researcher at Google, the difference in the inclusion of these languages was the use of a neural model of artificial intelligence that learned languages ”from scratch”.
Typically, implementing a new language requires millions of examples for the system to be able to “understand” and translate it. With neural models, additional languages were trained this way, according to Caswell. Technology then began to make sense of how languages worked.
The company says it consulted with representatives from several communities before releasing the new languages.
“Imagine you are a polyglot and based on your understanding of what languages are like, you can explain something. This is more or less how our neural network operates”, explained the researcher in a conversation with reporters.
First, 24 languages will be available for literal translation only. Google plans to add the voice part over time, allowing translation for anyone speaking or even facilitating the understanding of those who are curious about the sound of words.
Still speaking to reporters, Caswell said the languages would not be perfect. “The quality is lower than in English and Spanish. We know there will be some errors, but the tool will be useful,” he said.
Outside Latin America, Google added languages such as Kryo, a dialect of Sierra Leone English, Lingala (spoken by 45 million people in Central Africa, mostly in the Republic of the Congo), and Mizo, which is spoken by 800,000 people in northern India. spoken by.
News has been added to the more than 100 languages Google Translate already offers.
Below is the full list of languages added by Google Translate:
- Aymara – spoken by about 2 million people in Bolivia, Chile and Peru
- Assamese – spoken by about 25 million people in Northeast India
- Ashante – spoken by about 11 million people in Ghana
- Bambara – spoken by approximately 14 million people of Malik
- Boyapuri – spoken by about 50 million people in northern India, Nepal and Fiji
- Divi – spoken by about 300,000 people in Maldives
- Dogri – spoken by about 3 million people in North India
- JJ – spoken by 7 million people in Ghana and Togo
- Guarani – spoken by 7 million people in Paraguay, Bolivia, Argentina and Brazil
- Ilocano – spoken by about 10 million people in the northern Philippines
- Konkani – spoken by about 20 lakh people in central India
- Krio – spoken by about 4 million people in Sierra Leone
- Sorani Kurdish – spoken by about 8 million people (most of them from Iraq)
- Lingala is spoken by approximately 45 million people in the Republic of the Congo, Angola, the Republic of South Sudan, and the Central African Republic.
- Luganda – spoken by approximately 20 million people in Uganda and Rwanda
- Maitili – spoken by about 34 million people in North India
- Manipuri – spoken by 2 million people in Northeast India
- Mizo – spoken by about 830,000 people in Northeast India
- Oromo – spoken by 37 million people in Ethiopia and Kenya
- Quechua – spoken by 10 million people in areas close to Peru, Bolivia, Ecuador and countries
- Sanskrit – spoken by 20,000 people in India
- Sepedi – spoken by about 14 million people in South Africa
- Tigrinya – spoken by about 8 million people in Eritrea and Ethiopia
- Tsonga – spoken by approximately 7 million people in Eswatini, Mozambique, South Africa and Zimbabwe