HeyGen Developers (@HeyGenDev) / X

HeyGen Developers

172K posts

HeyGen Developers

@HeyGenDev

For developers building on @HeyGen and @Hyperframes_

developers.heygen.com

Joined April 2023

Following

27.9K

Followers

Replies Articles Media

Pinned
HeyGen Developers
@HeyGenDev
Jul 25
One-scene videos are good. Multi-scene videos in one API call? Deterministic output? Better. Happy with it? Turn it into a template and scale even further by just filling the variables. Studio API → developers.heygen.com/studio-videos Template API → developers.heygen.com/templates
00:00
HeyGen Developers
@HeyGenDev
Jul 27
The take was perfect. The audio wasn't. Swap the audio, keep the take. Lipsync API re-syncs the lips to match. One call. Precision mode. Lipsync API → developers.heygen.com/lipsync-precis… Batch → developers.heygen.com/batch-lipsyncs
00:00
HeyGen Developers
@HeyGenDev
Jul 27
New from HeyGen research: context engineering looks very different when your agent's output is a finished video instead of text. Our team wrote up what we learned shipping video agents -- how we manage multi-modal context, cut VLM costs, and ground models in time and space. Full
HeyGen Developers
@HeyGenDev
Jul 24
HeyGen research update: Multi-behavior avatar control without a co-labeled dataset. We train each behavior (expressiveness, gaze, camera) as a separate LoRA on single-attribute data, compose at inference, and close the train/inference gap with a short fine-tune on real overlap.
00:00
HeyGen Developers
@HeyGenDev
Jul 14
Article
Per-Word Timing for Real-Time Avatars
An avatar that acts while it speaks lives or dies on timing. The words are generated and voiced in real time, and the actions bound to them (a click, a scroll, a highlight) have to land on the word...
See @HeyGenDev's full profile
Sign up
Log in

See @HeyGenDev's full profile

See @HeyGenDev's full profile