The DigiU development speaks like a human being!

The DigiU.Voice service is based on a text-to-speech technology. This technology transforms text into speech with a personalised voice and individual intonation!

But just teaching artificial intelligence to talk is not enough. The speech must be as close to a human voice as possible. It means that AI must be able to reproduce intonations and emotions, copy speech, and pause in time for the DigiU.Voice service to be in demand.

Such a complex task was made possible by the DigiU professional IT team. Our data scientists have created a new architecture for building AI, adding additional layers, and applying new neural network learning mechanisms.

Due to this achievement, the DigiU.Voice service has successfully passed the patent search. We have already applied for a patent for DigiU.Voice as a "Method of cloning (synthesis) speech with the transfer of a reliable intonation of the cloned pattern" invention!

The Russian and English versions of DigiU.Voice are available without registration at https://lk.digiu.ai/voice. Check it out, share it with friends, and, of course, comment on the performance of our service.

To share the news, please enter your Personal Account!

Log in to your account