‘Ramayana: Rise of a Legend’: Namit Malhotra Says Human Actors Lead The Film’s AI-Assisted Dubbing Process

Producer Namit Malhotra has opened up about the use of AI in “Ramayana: Rise of a Legend,” explaining how the team is combining human performances with technology to create authentic multilingual versions of the film. The producer clarifies that voice performances are recorded by human actors, while specialists work on matching lip movements to the audio before AI is used to integrate the final output with the actors’ faces on screen.

Opening up about the usage of AI, Brahma AI to be precise, Malhotra tells ET Now, “The vivid example of that is what we created in our English trailer. We've got tremendous praise, but we've actually brought in human actors, voice actors who came in and performed the voice performances. Some of them actually (the dubbing) are done by the actors themselves who have a good command over the language and are able to emote appropriately in a different language. So we've done that as a first step to make sure that the rendering of that expression and emotion is authentic and led by humans. There is no AI whatsoever."

Related Stories

‘Ramayana’: Ranbir Kapoor, Yash and Sai Pallavi’s Film Gets Early Preview in France at Iconic Le Grand Rex on November 4

Inside Nitesh Tiwari’s ‘Ramayana’: Shiamak Davar’s Choreography, Songs, Part 2 Status and The World-Building (EXCLUSIVE)

He further adds, “Now we are trying to make the lip sync in line with the voice. So we have a next step, which is again, done by humans, where you actually have people coming in. They are what we call human drivers who create this foundation of how the lips move and how they sync themselves to the voice. And it looks appropriate because the best voice actor may not have the right face for the actor on screen.”

The producer quips that the work of AI comes after these human-led processes are complete. He elaborates, “We are making sure that you've got the voice actors. We’ve got people whose faces and lips match the look of the actors. And then we're using a bespoke technology tool that we've created, leveraging AI to stitch that into the face of the actor. So it feels absolutely seamless. So when you're now watching it, you feel like you're watching them actually speak with that authenticity and emotion, whilst not being taken away by subtitles and dubs that never quite deliver what's happening. So this is sort of how the technology was created to deliver at a cinematic quality and not just as something you can pull up in front of a phone and speak in multiple languages - that's much easily done. And that's not where you need this level of [expertise].”