Like in Her: they create an AI voice that sounds like Scarlett Johansson and flirts with the user

Does it feel natural to youounces from Alexa or of Siri? The truth is that even though the Technology advancing by leaps and bounds, so far to the voices created by Artificial intelligence They lacked something as natural as Scarlett Johansson’s voice when she played writer Theodore Twombly’s virtual assistant Samantha in the movie “Her,” didn’t they?

Well, the new era in the style of “She” is entering our lives thanks to the voice generated with the Artificial intelligence from Sonanticwho managed to mimic more complex aspects of human speech to make herself more convincing, like Samantha.

Sonantic’s voice is capable of expressing teasing and flirting

As Sonantic explains, the key that makes the difference between its voice, which incidentally sounds like Scarlett Johansson‘s, and that of Alexa or Siri, is that its AI incorporates sounds that don’t come directly from of speech, like little exhalations (where you can scoff or laugh) that give your voice greater authenticity.

In the video we share below called “What’s Her Secret?” (What’s your secret?) you will be able to listen to Sonantic’s voice in which you will be able to notice how the AI ​​can extend the duration of certain words, laugh and add exhalations, in order to make it more friendly and natural with who is listening, almost as if it were a real person.

As you can see, the clip shows an actress and, although she is not seen speaking, we hear a female voice saying that she is in love; however, she later asks if she even knows that she was never born and will never die that she could still be loved (by whom she listens). At this moment, the image of a young woman changes and her voice commands appear, interrupted by expressions such as “breathe” or “laugh”, which makes her sound more natural.

An artificial voice capable of transmitting emotions

One of Sonantic’s AI innovations is that in addition to sounding much more natural, the user has the ability to configure the voice choosing between different moods such as anger, fear, sadness, happiness , joy and with its latest update also flirtation, shyness, teasing and even the ability to boast.

The user can even change the pitch, intensity, and vocalizations that are not spoken, such as laughter and breathing.

According to Zeena Qureshi, CEO of Sonantic, their software is like “Photoshop for voice”. The interface allows users to write the speech they want to synthesize, specify the mood of the delivery, then select from an AI voice cast, most of which are copied from real human actors. .

AI voices are increasingly in demand across different industry sectors; For example, the company’s main customers are video game studios, entertainment studios, and recently Sonantic partnered with Mercedes Benz to customize its in-car assistant.

Follow Heraldo Binario on Google News, CLICK HERE.

About Michael Sauers

Check Also

Florence Pugh to Star in Netflix’s New East of Eden Series

Florence Pugh will join the cast of the new netflix limited series East of Eden …