15th March 2022 AI voices include subtleties of speech London-based company Sonantic has developed artificial voices that include subtle emotions and non-speech sounds.
In recent years, AI-generated speech has made significant advances. Male and female voices are becoming more and more lifelike and even include a range of different accents. However, they still lack the variety and natural quality of real humans and can often give away their robotic traits. Sonantic, founded in 2018, is developing voice software for film and gaming studios. For its latest project, it has produced a video in collaboration with a Hollywood client entitled "What's Her Secret?" The video surprises viewers when it's revealed that, while the woman on screen is a real person, the main character speaking is an AI. It aims to demonstrate how entertainment studios can leverage AI to create highly convincing romantic encounters. This marks a significant milestone for Sonantic as its technology is now able to recreate subtle emotions and non-speech sounds, while also opening up new creative possibilities for studios. The voice models, which already express a range of human emotions from happiness to sadness, can now convey subtleties such as flirty, coy, and teasing, amongst other new "Style" options. They also have the ability to capture non-speech sounds – such as breaths, scoffs, and laughs. This combination of advances in speech synthesis makes Sonantic's platform more comprehensive than ever before, helping entertainment studios create life-like performances in record time. "Human beings are incredibly complex by nature and our voices play a critical role in helping us connect with the world around us," said Zeena Qureshi, CEO. "Sonantic is committed to capturing the nuances of the human voice, and we're incredibly proud of these technological breakthroughs that we have brought to life through 'What's Her Secret?'. From flirting and giggling, to breathing and pausing, this is the most realistic romantic demo we've created to date, helping us inch closer to our vision of being the CGI of audio."
Comments »
If you enjoyed this article, please consider sharing it:
|
