Research, Technology

Artificial intelligence has learned to flirt and confess your love

On Valentine's Day 2022, the companySonantic posted a video that quickly went viral on social media. In the frame there is a girl who is silent and mysteriously smiles at the viewer. At the same time, in the background, a female voice flirts and confesses to the person watching the video in love. It would seem that this is an ordinary video, but at the end, the voice-over suddenly announces: “What you are hearing was never said by a person, it was generated by a computer. I'm not real. I was never born and will never die because I don't exist." It turned out that this is a demonstration of a new technology from Sonantic, which imitates a female voice using a neural network. He is so realistic that he knows how to flirt and confess his love. The technology is already being used in computer games and other areas.

Frame from a video published by Sonantic

The best speech synthesizer

The British company Sonantic has long been engaged indevelopment of synthetic voices. The first version of the technology, which mimics the human voice exactly, was developed in 2020. At that time, speech created by artificial intelligence did not sound natural enough, and distortion was heard in the background.

American actor Val Kilmer

Sonantic speech synthesizer in The Outer Worlds

The speech synthesizer learned to flirt

In 2022, the Sonantic voice synthesizer has become even morebetter. According to company representatives, in a few months they made the virtual voice more realistic by adding natural pauses and sighs between words - it is these details that make our speech "alive". Breathing, laughter and some timidity allowed the artificial voice to flirt and confess his love.

Without exaggeration, Sonantic has made a breakthrough in the field of speech synthesis

Take advantage of the technology of the momentmaybe not everyone. The synthesizer obviously works much better than analogues from Google and Yandex, therefore it is paid and available only by agreement of developers and interested customers. But representatives of Sonantic did not become too secretive and showed the interface of the utility, which is used to recreate the human voice. In the program, you can change the speed of pronunciation of words, adjust intonation, and add auxiliary sounds between words. In particular, we are talking about those very sighs, laughter and other subtle details of real human speech.

Sonantic speech synthesizer interface allows you to add auxiliary sounds

In the future, there will be programs for sounding films

Without any exaggeration, the technology is impressiveand intriguing. If one day it becomes available at an affordable price or has worthy analogues, there could be big changes in the field of voice-overs for cartoons and computer games. Almost everyone will be able to independently translate and release new series of their favorite animated series, and not wait for the day when voice acting with professional actors appears. If the technology becomes even more advanced, people with good diction and a beautiful voice may even be left without a job. In general, artificial intelligence is getting better and mastering more and more human skills.

Perhaps in the future, the profession of a dubbing actor will become less in demand.

If you are interested in the news of science and technology, subscribe to our Telegram channel. There you will find announcements of the latest news from our site!

What do you think of the new human speech synthesizer? Where else can it be used? Share your opinion in the comments or in our Telegram chat.