Replica Studios Simplifies Adding Synthetic Speech to Games and Movies
Voice AI startup Replica Studios has unveiled a desktop app to streamline integrating synthetic speech into films and video games. The new computer app makes it easier to connect the company’s ‘cast’ of AI voices to projects, including support for Unreal Engine and the new MetaHuman virtual human design platform.
Replica has created 40 AI voice actors based on recordings from human actors. The AI turns text into speech, with the user able to customize elements like the pace and the mood or emotion they want the AI to convey. The creator can then send the audio to incorporate into their video game or film project as a voiceover or as dialogue for characters, without the need for deep programming knowledge. As the project evolves, they can revise the AI audio whenever they choose. Though the AI voices are not a perfect imitation of human speech, the potential savings in money and time for edits and reshoots is an obvious attraction for film and video game directors. As part of the agreement to record their voices, the actors are paid with a piece of the revenue for any projects using their AI counterparts.
Unreal has been a particularly fruitful partner for Replica. Voices generated by Replica were used in making the animated film Cassini Logs, which won the Unreal Engine short film competition. Replica began beta testing a plugin for Unreal Engine in November, and developers have used it to record more than 16,000 audio sessions. In the wake of Unreal launching the much-hyped MetaHuman Creator and its hyperrealistic virtual humans in February, there are likely to be a lot more designers looking to combine the two platforms. The new desktop app will merge with the web app, which has been used to record more than 120,000 audio sessions in the last three months, according to the company.
“Our new Windows and macOS desktop app further offers accessibility to Replica’s high-quality AI character voices and makes
voice production significantly easier for games and filmmakers,” Replica CEO Shreyas Nivas said in a statement. “We think creators will love the breakthrough one-click export to Unreal Engine, Unity and support for Roblox Studio that our app offers and we’re looking forward to seeing the projects created with Replica voices – from games to animations, full-length films and more.”
Replica Studios raised a $2.5 million seed round at the end of 2019 to help complete and refine its platform, but it’s far from alone in the synthetic speech market. Startups like Resemble AI and Supertone are raising their own funding rounds to improve and market their synthetic voices, while digital voice interface creator ReadSpeaker recently agreed to augment SoundHound’s Houndify voice AI platform to sound more lifelike. The advances in synthetic speech technology mean even free, limited tools can produce impressive results like this fan-made trailer for Skryim voiced solely by AI. The tech giants aren’t ignoring the space either. Amazon’s voice synthesis tech has already turned Alexa into Samuel L. Jackson and given KFC Canada’s Alexa skill the voice of Colonel Sanders. Microsoft’s Custom Neural Voice service, meanwhile, has made AI versions of everyone from Bugs Bunny to Flo from Progressive Insurance, and Google is producing artificial voices for call centers.