Faceless Channel OS This workflow corresponds to n8n.io template #17261 — we link there as the canonical source.
The workflow JSON
Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →
{
"name": "Turn a text prompt into a free AI image and a Ken Burns zoom video clip",
"nodes": [
{
"id": "overview",
"name": "Overview",
"type": "n8n-nodes-base.stickyNote",
"position": [
-720,
-420
],
"parameters": {
"color": 1,
"width": 660,
"height": 1180,
"content": "## Turn a text prompt into a free AI image + a Ken Burns zoom video clip\n\nTypes a prompt, gets back a vertical (1080x1920) MP4: a free AI image slowly zoomed in \u2014 the motion faceless-video pipelines use so a still image doesn't look static on screen. No API key, no account, no credits anywhere in this workflow.\n\n**Pollinations** generates the image straight from a URL \u2014 no signup, no key. **FFmpeg's `zoompan` filter** animates a slow, continuous zoom over however many frames you ask for, turning a single AI image into something that reads as footage.\n\n### How it works\n\n1. A form takes your image prompt and how many seconds the clip should run.\n2. The prompt gets a few cinematic style words appended and is URL-encoded into a Pollinations image request.\n3. `curl` downloads the image to a per-run temp folder.\n4. `ffmpeg` crops it to 1080x1920, applies a continuous zoom sized to your requested duration, and encodes an MP4 \u2014 no audio track, since this is the visual half only.\n5. The result is checked for a clean exit code before you get the file paths back.\n\nDrop the MP4 into an editor, or feed it into a bigger pipeline that adds voiceover and captions on top.\n\n### Setup\n\n- `curl` and `ffmpeg` (with `libx264`) on the PATH of the process running n8n.\n- **Self-hosted n8n only.** This uses Execute Command, which n8n Cloud does not allow. n8n v2 also disables it by default \u2014 start n8n with `NODES_EXCLUDE=\"[]\"` and `N8N_RESTRICT_FILE_ACCESS_TO=\"/tmp\"` or the file steps fail silently.\n\n### Customization tips\n\nChange the zoom speed or ceiling by editing `zoom+0.0012` / `1.15` in the FFmpeg command. Swap in any other keyless image API by changing the URL built in the first Code node.\n"
},
"typeVersion": 1
},
{
"id": "section---input",
"name": "Section - input",
"type": "n8n-nodes-base.stickyNote",
"position": [
-40,
-180
],
"parameters": {
"color": 7,
"width": 460,
"height": 420,
"content": "## 1. Input\nYour image prompt and clip length. Style words are appended automatically."
},
"typeVersion": 1
},
{
"id": "section---fetch",
"name": "Section - fetch",
"type": "n8n-nodes-base.stickyNote",
"position": [
440,
-180
],
"parameters": {
"color": 7,
"width": 620,
"height": 420,
"content": "## 2. Generate the image\nPollinations builds the image straight from the URL. No key, no account."
},
"typeVersion": 1
},
{
"id": "section---render",
"name": "Section - render",
"type": "n8n-nodes-base.stickyNote",
"position": [
1080,
-180
],
"parameters": {
"color": 7,
"width": 460,
"height": 420,
"content": "## 3. Ken Burns zoom\nFFmpeg's zoompan animates a slow continuous zoom sized to your duration."
},
"typeVersion": 1
},
{
"id": "section---check",
"name": "Section - check",
"type": "n8n-nodes-base.stickyNote",
"position": [
1560,
-180
],
"parameters": {
"color": 7,
"width": 300,
"height": 420,
"content": "## 4. Verify\nConfirms FFmpeg exited clean before handing back the file paths."
},
"typeVersion": 1
},
{
"id": "note---self-hosted-only",
"name": "Note - self hosted only",
"type": "n8n-nodes-base.stickyNote",
"position": [
-40,
280
],
"parameters": {
"color": 3,
"width": 300,
"height": 160,
"content": "## Self-hosted only\nExecute Command does not exist on n8n Cloud. See Setup in the overview."
},
"typeVersion": 1
},
{
"id": "form-trigger",
"name": "Image prompt and length form",
"type": "n8n-nodes-base.formTrigger",
"notes": "Entry point. Everything downstream reads Image Prompt and Clip Length from here.",
"position": [
0,
0
],
"parameters": {
"options": {},
"formTitle": "Text prompt to AI image + Ken Burns zoom video",
"formFields": {
"values": [
{
"fieldType": "textarea",
"fieldLabel": "Image Prompt",
"placeholder": "A lighthouse on a cliff at sunset, storm clouds gathering",
"requiredField": true
},
{
"fieldType": "number",
"fieldLabel": "Clip Length (seconds)",
"placeholder": "5",
"requiredField": true
}
]
},
"formDescription": "Describe an image and pick a clip length. You get back a vertical MP4 with a slow zoom-in on a free AI-generated image."
},
"typeVersion": 2.2
},
{
"id": "build-paths",
"name": "Build the image prompt and file paths",
"type": "n8n-nodes-base.code",
"notes": "One /tmp folder per submission so runs never overwrite each other. Appends cinematic style words to the raw prompt.",
"position": [
220,
0
],
"parameters": {
"jsCode": "// One folder per run so two runs can never overwrite each other.\nconst runId = ($json.submittedAt || String(Date.now())).replace(/[^a-zA-Z0-9]/g, '');\nconst dir = `/tmp/kenburns-${runId}`;\n\nconst rawPrompt = ($json['Image Prompt'] || '').trim();\nif (!rawPrompt) throw new Error('The Image Prompt field is empty - nothing to generate.');\n\n// Pollinations reads the prompt straight from the URL path, so a few style words\n// appended here go a long way for a faceless-video look without any extra node.\nconst prompt = `${rawPrompt}, cinematic, dark moody lighting, vertical 9:16 composition`;\nconst imageUrl = `https://image.pollinations.ai/prompt/${encodeURIComponent(prompt)}?width=1080&height=1920&nologo=true`;\n\nconst seconds = Number($json['Clip Length (seconds)']) || 5;\nconst fps = 25;\n\nreturn {\n ...$json,\n dir,\n prompt,\n imageUrl,\n imagePath: `${dir}/image.jpg`,\n videoPath: `${dir}/clip.mp4`,\n seconds,\n fps,\n zoomFrames: Math.round(seconds * fps),\n};\n"
},
"typeVersion": 2
},
{
"id": "make-folder",
"name": "Create the temp folder",
"type": "n8n-nodes-base.executeCommand",
"notes": "mkdir -p is safe to run again.",
"position": [
440,
0
],
"parameters": {
"command": "=mkdir -p \"{{ $json.dir }}\""
},
"executeOnce": false,
"typeVersion": 1
},
{
"id": "download-image",
"name": "Download the AI image (Pollinations)",
"type": "n8n-nodes-base.executeCommand",
"notes": "Free, no account, no key. Retries a few times since Pollinations can be slow under load. Reads paths from the Code node, not $json: an Execute Command node replaces the item with its own stdout/stderr/exitCode.",
"position": [
660,
0
],
"parameters": {
"command": "=curl -fsS --retry 3 --retry-delay 5 --retry-all-errors --max-time 120 -o \"{{ $('Build the image prompt and file paths').item.json.imagePath }}\" \"{{ $('Build the image prompt and file paths').item.json.imageUrl }}\""
},
"executeOnce": false,
"typeVersion": 1
},
{
"id": "render-zoom",
"name": "Render the Ken Burns zoom video (FFmpeg)",
"type": "n8n-nodes-base.executeCommand",
"notes": "zoompan's d (frame count) is sized to your requested Clip Length so the zoom always finishes exactly at the end of the clip. Reads paths from the Code node, not $json, for the same reason as the download step.",
"position": [
1200,
0
],
"parameters": {
"command": "=ffmpeg -y -loop 1 -i \"{{ $('Build the image prompt and file paths').item.json.imagePath }}\" -t {{ $('Build the image prompt and file paths').item.json.seconds }} -vf \"scale=1080:1920:force_original_aspect_ratio=increase,crop=1080:1920,zoompan=z='min(zoom+0.0012,1.15)':d={{ $('Build the image prompt and file paths').item.json.zoomFrames }}:s=1080x1920:fps={{ $('Build the image prompt and file paths').item.json.fps }}\" -c:v libx264 -pix_fmt yuv420p \"{{ $('Build the image prompt and file paths').item.json.videoPath }}\""
},
"executeOnce": false,
"typeVersion": 1
},
{
"id": "check-render",
"name": "Verify the render succeeded",
"type": "n8n-nodes-base.code",
"notes": "Throws with the real FFmpeg stderr if the exit code was not 0, instead of silently returning a broken file path.",
"position": [
1620,
0
],
"parameters": {
"jsCode": "const cmdResult = $json;\nconst paths = $('Build the image prompt and file paths').item.json;\nif (cmdResult.exitCode !== undefined && cmdResult.exitCode !== 0) {\n throw new Error('FFmpeg failed: ' + (cmdResult.stderr || '').slice(-500));\n}\nreturn { json: { videoPath: paths.videoPath, imagePath: paths.imagePath, prompt: paths.prompt, seconds: paths.seconds } };\n"
},
"typeVersion": 2
}
],
"settings": {
"executionOrder": "v1"
},
"connections": {
"Create the temp folder": {
"main": [
[
{
"node": "Download the AI image (Pollinations)",
"type": "main",
"index": 0
}
]
]
},
"Image prompt and length form": {
"main": [
[
{
"node": "Build the image prompt and file paths",
"type": "main",
"index": 0
}
]
]
},
"Download the AI image (Pollinations)": {
"main": [
[
{
"node": "Render the Ken Burns zoom video (FFmpeg)",
"type": "main",
"index": 0
}
]
]
},
"Build the image prompt and file paths": {
"main": [
[
{
"node": "Create the temp folder",
"type": "main",
"index": 0
}
]
]
},
"Render the Ken Burns zoom video (FFmpeg)": {
"main": [
[
{
"node": "Verify the render succeeded",
"type": "main",
"index": 0
}
]
]
}
}
}
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
This self-hosted workflow collects a text prompt and clip length, generates a 1080×1920 image via Pollinations, then uses local FFmpeg to create a vertical MP4 with a Ken Burns-style slow zoom and returns the output file paths. Receives an image prompt and desired clip length…
Source: https://n8n.io/workflows/17261/ — original creator credit. Request a take-down →
Related workflows
Workflows that share integrations, category, or trigger type with this one. All free to copy and import.
This workflow contains community nodes that are only compatible with the self-hosted version of n8n.
Form Trigger, n8n, Form +2
Credentials Transfer. Uses form, httpRequest, executeCommand, readWriteFile. Event-driven trigger; 22 nodes.
Form, HTTP Request, Execute Command +2
This self-hosted workflow takes text and a selected language from an n8n form, generates an MP3 voiceover using the Edge TTS CLI, analyzes the audio with FFmpeg/FFprobe to detect real pauses, and outp
Form Trigger, Execute Command, Read Write File
ClaraAI - Pipeline V1 (Demo Call). Uses formTrigger, executeCommand. Event-driven trigger; 4 nodes.
Form Trigger, Execute Command
ClaraAI - Pipeline V2 (Onboarding Call). Uses formTrigger, executeCommand. Event-driven trigger; 4 nodes.
Form Trigger, Execute Command