Automation · Content and publishing

Create a promotion's image, video and voiceover on WhatsApp with Higgsfield

An AI agent uses Higgsfield to make the image, the video and the voiceover of the promotion you describe on WhatsApp, and sends them back.

Show me this with my own operation See every automation

How do I make the image and a voiced video for a promotion from WhatsApp, without editing?

Describe the promotion to your Met with Higgsfield connected: what is on offer, the tone, the format and what the voice should say. The Met checks which model fits each piece, generates the image, shows it to you and, once you approve it, generates a short video from it and the voiceover. If you prefer a ready-made format, such as a faceless video or a UGC style one, it loads the production workflow Higgsfield publishes and follows it. Everything arrives in the conversation. Every generation uses your Higgsfield account's credits, which is why the image, video and audio generation actions start switched off: the admin has to turn them on in Connections before using this recipe.

What starts it

When someone asks for it

It starts when someone at the business describes the promotion to the Met on WhatsApp and asks for the pieces. It does not run in the background: each piece is a request and uses Higgsfield credits.

A manual task: someone on the team runs it with a button, or asks the Met for it in chat.

Who runs it

Content Met that builds promotion pieces with Higgsfield

Studio team

This is not a named, seeded Met. It is an internal Met with Higgsfield connected: its work ends when the pieces reach whoever asked for them, not when they are posted to social media.

Run by the social video Met

Channels: WhatsApp

What you need connected

Higgsfield Create and promote

5 steps · 5 tools

The procedure, step by step

Each step shows the tool that runs. They belong to the real connector: if one of them stopped existing, this page would not build.

1
Choose the model for each piece

Before generating, the Met explores the models available on your account and their parameters, for the image, the video and the audio. That way it does not make up a model or a setting Higgsfield does not accept.
- models_explore Higgsfield
2
Generate the promotion image

With your description and the format you asked for, the Met generates the image and shares it in the chat. If you do not like it, you tell it what to change and it generates another; each attempt uses credits. This action starts switched off: the admin has to turn it on.
- generate_image Higgsfield
3
Turn it into a short video

With the image approved, the Met generates a short video by animating it, with the motion, aspect ratio and duration you asked for. It shares it once Higgsfield finishes it. This action starts switched off: the admin has to turn it on.
- generate_video Higgsfield
4
Generate the voiceover

With the text the voice should say, the Met generates the voiceover audio and shares it separately, so you can lay it over the video or use it in a voice note. This action starts switched off: the admin has to turn it on.
- generate_audio Higgsfield
5
Use a ready-made workflow if you prefer

If instead of separate pieces you want a complete format, such as a faceless video or a UGC style one, the Met checks Higgsfield's catalog of production workflows, loads the instructions of the right one and follows them.
- get_workflow_instructions Higgsfield

Before you start

What you need ready

The Met solves none of these for you. If one is missing, the automation stops there.

A Higgsfield account with credits

Your business's Higgsfield account, with credits to generate. Usage runs on your account; Meteor does not resell Higgsfield.
Higgsfield connected in Met

In Met, open Connections → Higgsfield → Connect and sign in with your account. The image, video and audio generation actions start switched off because they spend credits; this recipe needs the admin to turn them on there.
A WhatsApp line with the Met assigned to the team

The business connects its WhatsApp Business API line and assigns the Met that serves the team. The request and the pieces travel in that conversation.

The limits

What this automation does NOT solve

Better said here than discovered halfway through the rollout.

It does not use the real photo of your product: this recipe starts from a description. If the piece has to show your product exactly as it is, it is best made with its photo in the Higgsfield app.
It does not post the pieces to your social networks. It hands them to you in the chat; posting them is another step, with another integration or by hand.
It does not ask for confirmation in the chat before generating: once the admin turns on a generate action, the Met generates and uses credits. The control is which actions the admin leaves on.
It does not merge the video and the voiceover into one file when it generates them separately; for a joined piece, use a Higgsfield production workflow.

FAQ

Questions about this automation

The sum of what Higgsfield charges for each generation: the image, the video, the audio and every extra attempt. It depends on the model and the settings, just like in its app.

Yes. Tell the Met the aspect ratio and duration you need; it passes them as conditions when generating the video.

Yes, if the audio model the Met picks supports it. That is why it checks the models and their parameters before generating.

Keep going