Google Translate started as a fairly basic translation tool that was put to use only if there wasn’t any other option, but over the years, the tool has come a long way.
Translate’s AI-backed features are invaluable for translating everything from a few sentences to entire web pages, with a reliable degree of accuracy and readability.
Google’s stepping that up with the unveiling of Translatotron — an “experimental new system” that Google says will translate speech directly into speech, removing the need for any text.
“Translatotron is the first end-to-end model that can directly translate speech from one language into speech in another language,” says Google in an AI blog post.
Google says that there are three stages of today’s translation systems — automatic speech recognition, which transcribes speech as text; machine translation, which translates this text into another language; and text-to-speech synthesis, which uses this text to generate speech.
Cascading these steps led to the creation of services like Google Translate, but the tech giant now says it will use a single model without the need for text.
Translatotron also avoids dividing the task into separate stages which essentially means faster translation speed and less compounding errors, according to Google. The system is also able to retain the speaker’s vocal characteristics once translated.
The inner workings of Translatotron is fairly technical in nature, but in simpler terms, you should be able to say something like “My name is Ronny, how are you?” in English to someone who only speaks Spanish and have the tool convey accurately and quicker than ever before. You can head here and try out some of the audio samples shared by Google
However, Translatotron isn’t available for the general public just yet, and there’s no word on when it might launch. Does that mean that we’ll perhaps see it being quietly integrated into Google Translate itself in the future? Our guess is as good as yours.