Model · HeyGen

HeyGen Video and HeyGen Lipsync, no HeyGen account needed

On Beemm Vision, HeyGen means two models: HeyGen Video, which generates a video with sound from text, an image or references, and HeyGen Lipsync, which re-syncs the lips of an existing video to a new voice. Both are picked in Generate, under HeyGen, or on a canvas node, on the same credit balance as the rest of the catalogue. This is not the HeyGen avatar app: you bring your own images, video and voice.

What HeyGen brings

A video that comes out with its sound

HeyGen Video generates the picture and the soundtrack together, with no switch to turn on: the clip arrives with sound, 5 to 15 seconds long, at 480p, 768p, 1080p or 2K. 768p is the default.

Three ways to start

Text alone; a start image, which becomes the first frame of the shot and sets its framing; or references: up to nine images, three videos and three audio tracks, which the prompt cites as Image 1, Video 1, Audio 1. These are the Frames, Reference and Text modes of the picker.

Dub a video onto a new voice

HeyGen Lipsync takes a 1 to 30 second video and a WAV or M4A voice, shorter than the video, then re-syncs the lips to it. The video keeps its length. Two tiers: Speed, the default, or Precision. The step-by-step, and the other models that dub a video or make a portrait talk, are in the lip-sync guide.

The cost measured before you pay

HeyGen Lipsync is billed on the exact length of the video, measured before anything is charged: Generate shows the amount as soon as the file is measured. An MP3, a file outside the bounds or a voice longer than the video is refused before any charge.

The two HeyGen models available here

Filed under HeyGen in the Generate video picker, and available on the canvas.

  • HeyGen VideoText, start image or references (up to 9 images, 3 videos and 3 audio tracks) to video with sound. 5 to 15 s, at 480p, 768p (default), 1080p or 2K.
  • HeyGen LipsyncA 1 to 30 s MP4 or MOV video and a shorter WAV or M4A voice: the lips follow the new voice, the video keeps its length. Speed or Precision. Reserved for accounts declared adult.

What Beemm Vision adds

A video is prepared upstream and reworked downstream, in the same graph.

  • Prepare the start image or the references with Nano Banana or Seedream, then animate them with HeyGen Video, without an export.
  • Compare HeyGen Video, MiniMax H3, Kling and Seedance on the same brief before choosing.
  • Wire the generated video into the Video input of a HeyGen Lipsync node to dub it, without downloading it.
  • Run HeyGen Video from Claude or ChatGPT through the MCP connector.
  • One credit balance, and no HeyGen API key to manage.

HeyGen on Beemm Vision: what is here, what is not

A generation model, not an avatar

HeyGen also makes a talking-avatar app. HeyGen Video is something else: a video generation model which, according to its listing on FAL, which runs it, is built on MiniMax H3 and post-trained by HeyGen. It starts from your prompt and your media, not from an avatar library. To make a portrait talk over your voice, use the Avatar-type models — Kling AI Avatar v2, MiniMax H3 Max Lip Sync, OmniHuman v2 or sync-3 Avatar — covered in the lip-sync guide.

Writing for HeyGen Video

A prompt is required in all three modes, start image included. In Reference mode, cite each medium by its rank — Image 1, Video 1, Audio 1 — to say which one holds the character, which the set, which the rhythm; a reference audio never goes alone, it needs at least one image or video. In Text mode, six formats from 21:9 to 9:16, plus an adaptive format in Reference mode; in Frames mode, the image sets the framing. On the canvas, the node panel also sets the automatic prompt enhancement, on four levels including one that turns it off, and the seed.

Frequently asked questions

What is HeyGen on Beemm Vision?
Two models from the HeyGen company: HeyGen Video, which generates a video with sound from text, an image or references, and HeyGen Lipsync, which re-syncs the lips of an existing video to a new voice. Pick them in the video tab of Generate or on a canvas node.
What length and resolution does HeyGen Video offer?
5 to 15 seconds, picked to the second, at 480p, 768p, 1080p or 2K; 768p is the default. Sound is always generated with the picture.
How does HeyGen Lipsync work?
In Generate, pick HeyGen Lipsync, then drop a 1 to 30 second MP4 or MOV video into "Video to dub · 30 s max" and a WAV or M4A voice, shorter than the video, into "Voice · WAV or M4A". The model re-syncs the lips to that voice and the video keeps its length. Two tiers: Speed, the default, or Precision. No prompt to write.
Does HeyGen Lipsync translate a video?
No. It re-syncs the lips to the audio you supply: record the translated version first, as WAV or M4A, shorter than the video. An MP3 is refused before any charge, so convert it first.
Is HeyGen open to every account?
HeyGen Video, yes. HeyGen Lipsync has no content filter on its provider’s side: it is reserved for accounts that have declared they are adults. Other accounts can dub a video with sync-3 Lipsync or Kling LipSync.
Do I need a HeyGen account?
No. Beemm Vision calls the model for you and bills in credits on a single balance shared across the whole catalogue.

Try HeyGen Video

The button opens Generate with HeyGen Video already selected. The full catalogue is available from the free plan.

Keep reading