YouTube has introduced updates to its auto dubbing system aimed at improving how multilingual videos are experienced by both viewers and creators. The changes expand language support, add realism to translated audio, introduce viewer controls, and improve creator-side handling of dubbed content, said Chandralekha Motati, Product Manager, YouTube.
Expanded language support and expressive speech
Auto dubbing now supports 27 languages and is available across YouTube. According to the company, in December more than 6 million viewers watched at least 10 minutes of auto-dubbed videos daily on average, reflecting growing cross-language viewing.
YouTube has also added Expressive Speech, a feature designed to retain tone and emotional delivery from the original audio when translated. This is intended to make dubbed speech sound closer to the creator’s natural voice style.
Viewer language preferences
A new Preferred Language setting gives viewers direct control over how dubbed audio is played. While YouTube already selects a playback language based on watch history, users can now manually choose their preferred language or select the original audio when available.
Lip sync pilot for visual consistency
YouTube is testing a Lip Sync pilot that adjusts a speaker’s lip movements to better match translated audio tracks. The goal is to make dubbed videos appear more visually aligned with spoken dialogue.
Creator controls and filtering improvements
Several updates focus on how auto dubbing works for creators:
- Automatic Smart Filtering: Detects videos that may not be appropriate for dubbing, such as music-focused or silent content, helping prevent mismatched audio.
- Discovery behavior: YouTube states that auto dubbing does not negatively affect a video’s original discovery performance and may support reach in additional languages.
- Manual control: Creators can upload their own dubbed audio tracks or disable auto dubbing entirely.
Availability
- Auto dubbing is available to YouTube creators with support for 27 languages.
- Expressive Speech is supported in eight languages — English, French, German, Hindi, Indonesian, Italian, Portuguese, and Spanish.
- The Preferred Language viewer control is available to multilingual audiences.
- The Lip Sync feature is currently being evaluated through a limited pilot.