Generate AI UGC product videos from a Telegram photo with OpenAI

Find Your AI Training Job
Chat with DigiNo Toucan and get matched to remote roles in one minute.

Ecommerce sellers need constant ad creative but can't afford actors or video crews. This AI automation takes a product photo sent via Telegram and returns a finished UGC-style video ad, making it a repeatable service you can sell to Shopify stores and Amazon sellers.

What This Automation Does

Accepts a product photo and brief text prompt through a Telegram bot, then analyses the image to extract product details, brand identity, and optional character information
Uses OpenAI to generate a structured creative brief and matching image prompts, then produces AI-generated visuals tailored to the product and character
Chains the generated images into short video clips and combines them into a single 15–30 second UGC-style ad ready for paid social campaigns
Delivers the finished video back to the sender via Telegram or routes it to cloud storage, completing the full production loop without any manual editing

Tools Used

n8n
OpenAI
Telegram

Work faster

Stop Typing and Start Speaking With Typeless

Typeless turns what you say into clean, formatted text in any app on Mac, Windows, Linux, iOS and Android, which it rates at 4x faster than typing. It strips filler words, keeps only your final phrasing when you change your mind mid-sentence, and works in 100+ languages. Emails, AI prompts, client messages and docs, done by voice.

Try Typeless →

Where to Get Hired for This Skill

On Contra, top freelancers across this stack have earned 102 combined verified reviews from real client projects.

Source: Contra freelancer search · refreshed 30 May 2026

Start Earning as a Freelancer on Contra

Contra is a commission-free professional network for independents. Browse live AI automation work and keep what you earn.

Join Contra Free →

How To Build It

Wire Telegram to receive product photos

A Telegram bot listens for incoming messages containing an image and a text prompt, capturing both the compressed product photo and the user's creative instructions as the workflow's starting input.

Extract product and character data with OpenAI

The uploaded image is retrieved and passed to an OpenAI vision model that identifies the product's brand, colour, and description alongside any character name, outfit, and visual style present in the photo.

Generate structured prompts for each video scene

OpenAI produces a set of scene-by-scene image generation prompts derived from the extracted product and character data, with each prompt mapped to a specific clip segment of the intended video duration.

Create and sequence the AI-generated visual clips

Each prompt is sent to an image generation model in parallel, the resulting images are converted into short video clips, and those clips are aggregated and stitched into a single continuous video file in the correct running order.

Deliver the finished video to the client

The compiled video is sent back to the original Telegram conversation as confirmation of completion, with an optional parallel route that saves the file to Google Drive or Dropbox for client handoff and archiving.

Pitfalls

Image generation APIs enforce per-minute and per-day rate limits that cause silent failures mid-workflow when processing multiple scenes in parallel, so you need explicit wait and retry logic between clip generation calls to avoid incomplete videos.
Compressed photos sent through Telegram often lose enough quality to confuse the vision model's character and product extraction, producing vague or incorrect prompts that result in off-brand output, so clients must be given a clear image spec before they submit anything.
Video clip stitching depends on consistent resolution and frame rate across all generated images, and any mismatch in the output from the image model will break the final assembly step, requiring a normalisation pass before the clips are combined.

FAQ

Can I build this without coding?

Yes. The entire workflow runs in n8n using its visual builder and pre-built integrations for Telegram and OpenAI. The only configuration that requires reading documentation is setting up the Telegram bot token and the OpenAI API key, both of which involve copying and pasting credentials rather than writing code.

How long does it take?

Initial setup and testing typically takes a few hours spread across a day or two. End-to-end video generation for a single product once the workflow is live takes roughly 10 to 20 minutes depending on the number of scenes and API response times.

What can I charge?

Pricing is entirely up to you based on your market research and the value you provide to clients. Many freelancers offering AI video production services package deliverables by output volume or turnaround speed rather than charging hourly.

Which tool is required vs optional?

OpenAI is required for both image analysis and prompt generation, and Telegram is required as the input and delivery channel in the default setup. Cloud storage destinations like Google Drive or Dropbox are optional delivery routes you can add if clients prefer file-based handoff over Telegram.

This is original DigiNo analysis. The underlying automation pattern is a community workflow template – view the original on n8n.