Turn short videos into clear voiceovers with one run. The automation pulls a video, writes a simple script from its frames, and produces audio you can store and share from Google Drive. Great for content teams, social clips, product demos, and quick explainers.

A manual test starts by downloading a video from a URL. Python with OpenCV captures up to 90 evenly spaced frames from the clip. Frames are split into groups of 15, resized, and sent to an OpenAI model that can read images to draft parts of the narration. Each round adds to the story so the script stays consistent. A wait step helps avoid rate limits. The final script goes to OpenAI text to speech, and the mp3 file is uploaded to Google Drive.

You will need an OpenAI API key and a Google Drive account. Keep videos small or limit frame count to avoid high memory use. Expect to reduce manual scripting and voice recording from hours to minutes for short clips. This setup fits teams that need frequent narrated content without studio work.

• Manual test trigger to control when processing starts
• HTTP download of the source video from a URL
• Python Code node with OpenCV to capture up to 90 evenly spaced frames
• Split frames into batches of 15 to manage token limits
• Convert and resize frames before sending them to the model
• OpenAI multimodal prompt to draft narration from images
• Wait step to respect service rate limits for stable runs
• Aggregate partial outputs into a single clean script
• OpenAI text to speech to create an mp3 voiceover
• Google Drive upload with a timestamped file name

• Reduce manual scripting and recording from 2 hours to about 10 minutes per short clip
• Automate up to 90 percent of narration work by batching frames and carrying context forward
• Stay within model and rate limits using built in batching and a wait step
• Improve script consistency by combining partial drafts into one final version
• Deliver finished mp3 files directly to Google Drive for easy sharing
• Scale output by adjusting frame count and batch size for longer videos

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with OpenAI and Google Drive. See the Tools Required section above for links to create accounts with these services.
3. Open the HTTP Request node and set the video URL you want to process. Use a small clip while testing.
4. Double click the Code node named Capture Frames. Keep the default Python script if you want up to 90 frames. Lower the max frames value if you run into memory issues.
5. Open the Split In Batches node and confirm the batch size is 15. Adjust this if you change the total frame count so batches stay balanced.
6. Check the Convert to Binary and Resize Frame nodes to confirm the image property names match the output from the Code node.
7. Open the OpenAI Chat Model and Generate Narration Script nodes. In the credential dropdown, click Create new credential, select OpenAI, add your API key, and save.
8. Open the Use Text-to-Speech node. In the credential dropdown, select the same OpenAI credential. Choose a model and voice that fit your brand tone.
9. Open the Google Drive node. In the credential dropdown, click Create new credential and sign in with Google to allow file upload. Choose the target folder.
10. Keep the Stay Within Service Limits node enabled if you see rate limit errors. If your account has higher limits, you can disable it for faster runs.
11. Run a manual test from the top. Watch the Combine Script node output to confirm the full narration text looks right. Then check Google Drive for the new mp3 file.
12. If you get memory or timeouts, reduce frame count, shorten the video, or increase the wait time. For Drive permission errors, reconnect the Google Drive credential.