antitrust investigation

We all know that communication is necessary to convey your message but more importantly, language is the key to start that process of communication. That is why Google has started working on the latest prototype AI translator which can speak in another language using your own words. It does not just translate the words coming out of your mouth but also the tone and cadence of your voice.

The system is called Translatotron, in a recent blog post, Google has explained every bit of information. They don’t say that Translatotron products will available soon commercially but it might be possible to release anytime soon. Google head of translation said the company’s aim is to add more nuance to its translation tools, creating more lifelike speech.

The AI translator is described as an end-to-end, speech-to-speech translation model. What makes it more innovative that instead of using the usual method of speech-to-text and then text-to-speech voice conversion, it uses a neural network so it can skip the intermediate step of translating audio to text and back again. Plus it now has a speaker encoder which will preserve the original voice of the person. Google says that making translation end-to-end has more benefits like it produces results faster whereas avoiding the risk of introducing errors during multiple translation steps.

Also, the data model is not processing raw audio but it uses spectrogram data or detailed visualizations of sound. It means we are translating speech from one language to another by using only pictures.

read also: Apple’s new TV app is still not so good for watching TV on iPhones

Google further said Translation quality is you hear does not like previous translation rather it sounds more natural because they are not using the conventional methods. Currently, there have been various translation apps present in the market like Google Translate, SayHi, Microsoft Translator, iTranslate, and TripLingo.

But none of the above apps use your real voice in the translation. As far as Google efforts for making the translation better is concerned, we are not sure how it will work in the field. Although the technology is getting better day by day so we should hope for the best.