Artificial intelligence is learning to do things that, until a few months ago, seemed like something out of science fiction. However, these uses are already with us, and in today’s post we bring you a new AI tool that you can use from your iPhone and that translates your voice. Yes, however strange a language may seem to you, Thanks to this algorithm, you will speak it (on video) just as if you were a native speaker.
Create a deepfake of your voice, but in another language
The recreations of our person, in its different aspects, is already manifesting itself in different ways in the world of artificial intelligence. Whether it’s putting our face on another person or giving our voice to processors so we can recreate it virtually, this has only just begun.
And one of the iterations that has recently come to light is Rask.AI, a web tool that we can open from Safari, which has a very simple premise: we record ourselves on video, speaking in front of the camera; we upload the video to the web application and wait for it to process and, finally, we will receive the same video, with our face and our voice, but translated in the language that we want. Of course, we must have that we are talking about a newly created tool, in which we are using the free version. This implies that the languages that are more similar to each other or that are more commonly translated will have a much more accurate result, as far as our voice is concerned. But this is AI, so it’s also a matter of a short time to see a significant improvement in that aspect.
It seems hard to believe at first, but the truth is that it is a tool that works surprisingly well for being so recently created. At the moment, only the section of translating our voice works. But there is no lip sync. That is, that in future versions, the developers have already said that, in addition to the translated voice, this will be synchronized with the diction and the movement of our lips.
How our voice is translated in Rask.AI
The operation is very simple. Once we have logged in, we will have a panel in which we will see the projects that we have translated. All we have to do is select the video we want, adjust the language parameters and let the AI process it. And how does he do it? Well, by transcribing what we say in the video, to later translate it and apply it with the voice deepfake technique.
When the video is translated, the website tells us that we can close the page. And in a few minutes, we will receive an email with the project already translated. But we can also obtain the result of our translation in the Rask.AI web interface itself.
Rask.AI has a free plan with which we have access to translate two videos per month, maximum 100 mb in size and one minute long. However, with the payment plan, this limitation is unlocked. And it is that we will have up to 25 minutes of maximum duration per video, in more than 60 different languages, in the plan of 39 dollars per month. But if we want to go further, we can go for the Pro plan, which costs $99 per month and with which we will have early access to the lip sync function, and a maximum length of 100 minutes for each video.
