Today we’re announcing our new lip sync model: RealFace. RealFace is the result of years of research and months of focused engineering, and we’re proud to finally share it. Most AI video tools generate synthetic people. We built RealFace to adapt real ones—preserving expression, lip movement, and performance even when the language changes. It holds up under jump cuts, angle shifts, and natural speaker movement, because that's how real video works. See what makes RealFace different. 👇 Want to learn more? See the link in the comments.

60% of Spotify’s podcast listeners are non-English speakers. If you're a podcast creator, that's huge opportunity for growth. That's why we just created Podcast Pro, our podcast localization platform. Key Features: ✔️ Multi-Speaker Support: Accurate lip, gesture, and body sync across different voices. ✔️ AI-Driven Speech Modulation: Retains your authentic tone, rhythm, and emotional nuances. ✔️ Freedom from Occlusions: No re-recording needed due to hand or mic interference. ✔️ Contextual Adaptation: Captures natural speech patterns: breaths, pauses, and all. ✔️ Scalable Localization: Translate single episodes or entire libraries effortlessly. ✔️ Human-in-the-Loop Tools: Precision editing for perfect alignment with your brand’s voice.