Generate and upload ASMR YouTube Shorts daily with GPT-4

See What's Earning in AI Automation Freelancing.
DigiNo helps new AI automation freelancers earn faster by tracking what clients actually pay for.

YouTube automation clients want a channel that publishes itself, but building a full video pipeline from scratch takes weeks. This AI automation handles ideation, video generation, audio creation, and YouTube publishing in a single scheduled workflow.

What This Automation Does

Generates a trending ASMR concept daily using a two-stage AI planning process that produces a structured production brief, caption, and hashtags
Creates a matching video clip by feeding scene prompts to a text-to-video model, then generates layered ASMR audio via a separate text-to-audio model
Combines and publishes the finished Short directly to a YouTube channel on a fully automated schedule without any manual intervention
Logs each published video to Google Sheets and sends a confirmation notification via Telegram or Gmail so clients can monitor output remotely

Tools Used

n8n
OpenAI
Google Sheets
Gmail
Telegram
YouTube

Where to Get Hired for This Skill

On Contra, top freelancers across this stack have earned 194 combined verified reviews from real client projects.

Source: Contra freelancer search · refreshed 26 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 up the daily schedule trigger

Set the workflow to fire at a fixed time each day so every subsequent step runs automatically without anyone touching a keyboard.

Build the two-agent ideation chain

Configure a first AI agent to brainstorm a short viral ASMR concept, then pass that concept to a second planning agent that outputs a structured brief containing scene descriptions, sound cues, a caption, and hashtags.

Connect video and audio generation APIs

Route the scene description from the planning brief to the Seedance text-to-video endpoint via Wavespeed AI, and simultaneously send the sound cue description to the Fal AI text-to-audio endpoint, using wait steps to hold execution until both assets are ready.

Authenticate and publish to YouTube

Link a client YouTube account using OAuth, then pass the generated video file, AI-written caption, and hashtags directly to the YouTube upload step so the Short appears on the channel fully titled and described.

Log output and send client notifications

Write each published video's title, URL, and timestamp to a Google Sheets log the client can monitor, then fire a Telegram or Gmail message confirming the upload so the client has a daily paper trail without logging into YouTube.

Pitfalls

Seedance and Fal AI generation times vary under load, and if wait durations are set too short the workflow will attempt to download an asset that is not yet ready, breaking the upload step silently.
YouTube OAuth tokens expire and must be refreshed, and if the client rotates their Google account password or revokes app access the entire pipeline stops publishing with no visible error until someone audits the log.
The planning agent can drift toward repetitive ASMR concepts after a few weeks if the prompt lacks variety instructions, resulting in near-identical videos that depress channel performance and prompt client complaints about deliverable quality.

FAQ

Can I build this without coding?

Almost entirely, yes. The workflow is low-code and relies on pre-built service integrations for YouTube, OpenAI, and the notification tools. The only parts that may need a short code block are combining the video and audio assets before upload, which requires basic familiarity with handling binary data in n8n.

How long does it take?

Expect two to four hours for an initial build and test run from scratch. A second delivery for a different client reusing the same structure typically takes under an hour since the core logic is already proven.

What can I charge?

Pricing is your decision based on your market and positioning. A common approach is a one-time setup fee for delivery plus a smaller recurring monthly fee covering maintenance, token costs, and API monitoring, which gives clients a predictable expense and gives you stable income.

Which tool is required vs optional?

OpenAI, Seedance via Wavespeed AI, Fal AI, and YouTube are all required as the core generation and publishing stack. Google Sheets for logging and Telegram or Gmail for notifications are optional quality-of-life additions that most clients appreciate but are not needed for the pipeline to function.

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