Google Translate goes a step further with ‘Translatotron’, a service that translates without the need for text. The translator converts your audio to another language, preserving the tone of your voice.
The Google company has announced this new service that will be carried out through Artificial Intelligence mechanisms. The project, which is still in the test phase, processes the audio from source sequence to sequence in spectrograms and generates other new audio models with translated content and then converts it into the target language.
The service encodes the characteristics of the voice in order to maintain the tone of the translation.
“This system provides some advantages over other solutions, including greater speed and avoiding compositional errors between recognition and translation, which facilitates the retention of the original speaker’s voice after translation and better handling of words that do not need to be translated, for example, proper names,” said company sources.
Google takes a step forward with Translatotron’, which offers a personalized service and, according to the company, with fewer errors. “It is the first model that can directly translate spoken words in one language into spoken words in another,” said the company.