At Kapwing, we help marketing and communications teams build engaging content fast. We’ve always aimed to make video creation easier, smarter, and more accessible. That’s why we’re thrilled to announce our latest partnership with Sieve, a leading AI infrastructure company focused on video data. Sieve is one of the vendors backing our newest AI Persona feature, which launches today.

What is Sieve?

We met Mokshith, the Sieve CEO, years ago when he and his cofounder were exploring bounding box technology for video companies. Our CTO partnered with the then-tiny YCombinator startup on an object detection feature. We’ve had a front-row seat to how they’ve grown their business to support many leading video products and have been impressed with their focus on developer experience and relationships. 

Sieve offers a lip sync pipeline that uses artificial intelligence to match a video’s lip movement to audio. At Kapwing, we leverage this technology in our Personas feature, which enables you to make a video without recording any footage, as well as our Dubbing workflow. Sieve makes it easy to ensure that the lips on screen match perfectly with the audio being said.

Through our integration with Sieve’s platform, Kapwing users can now automate the lip-syncing process, eliminating the need for painstaking manual adjustments. Here's what this means for you:

• Less recording, more content: Record a single clip of video footage, and reuse it to say any message you’d like. With Sieve’s lip sync, matching existing footage to new speech happens in minutes, enabling you to scale your content in new ways.
• Global reach: Easily dub videos in multiple languages while maintaining natural lip movements, helping you tap into new audiences.
• Increased creativity: Kapwing’s video editor has Personas embedded so that you can add voice overs, collaborate with influencers, or make conversations with your AI twin.

Why Sieve?

Sieve’s team has been at the cutting edge of video AI models, and their investment and expertise with lip sync has been no exception. As one of the first providers to publish an eval on open-source models, we knew that they were familiar with all the latest developments in the space. Thus, when MuseTalk released in April of this year, we knew we had to chat with the Sieve team to learn what they had planned, and optimizations that they could make to the latest developments in lip sync.

Partnering with Sieve gives creators on Kapwing more choices and control over the models that run on their videos, and allows them to explore the possibilities for using AI in their video creation process. With this partnership, we hope to empower even more creators with AI-driven tools to produce polished, professional videos, no matter the language or format.

We’re excited to see how our community uses this new feature to push their creativity and connect with audiences across the globe. Thousands of videos have already been created using AI Personas on Kapwing, and hundreds of creators have tried custom personas and leveraged lip sync to generate video. With this partnership, we’re excited to see what else gets created. 

Stay tuned for more updates, and be sure to try out Sieve’s lip sync tool in your next Kapwing project. Happy editing!