YouTube Expands Auto-Dubbing to 27 Languages

YouTube has expanded its auto-dubbing feature to 27 languages and introduced new AI features such as expressive speech and lip-sync testing.

The company said auto dubbing is now available to all creators, a scale-up from earlier limited access. In December alone, YouTube recorded an average of more than 6 million daily viewers who watched at least 10 minutes of auto-dubbed content, underscoring growing viewer comfort with AI-generated translations.

YouTube uses generative AI models to make dubbed videos sound and feel more human. The platform has rolled out Expressive Speech across eight languages, such as English, French, German, Hindi, Indonesian, Italian, Portuguese and Spanish, designed to capture a creator’s original tone, emotion and delivery.

Unlike traditional text-to-speech systems that prioritise accuracy alone, the AI models powering expressive speech attempt to mirror pacing, emphasis and emotional cues, addressing long-standing criticism that dubbed content feels robotic.

YouTube is also experimenting with AI-powered lip-sync technology, currently in pilot testing. The feature subtly adjusts a speaker’s lip movements to match translated audio, reducing visual dissonance and making dubbed videos appear closer to native-language content. 

If scaled widely, the technology could reshape how international audiences consume video, particularly in education, explainers and creator-led storytelling.

A new Preferred Language setting allows users to choose whether they want videos to play in a selected language or retain the original audio, adding transparency and agency to YouTube’s AI-driven localisation tools. This builds on the platform’s existing system that automatically selects languages based on watch history.

For creators, YouTube says AI dubbing is designed to be “all gain and no pain.” New Automatic Smart Filtering, powered by machine learning, identifies videos that should not be dubbed,such as music videos or silent vlogs,to preserve creative intent. 

The company also said auto dubbing does not negatively affect discovery and may help content surface to audiences in new geographies.