🎬 Create captioned podcast audiogram clips with Zvid, ElevenLabs, and OpenRouter
⚡ 1 views · 🎬 Content Creation & Video
Description
Quick overview
For podcasters and content teams: turn the newest RSS episode into captioned vertical audiogram clips using ElevenLabs transcription, OpenRouter highlight selection and Zvid rendering.
How it works
- Run manually or daily to select the newest audio episode from your RSS feed.
- Download a bounded sample and check its actual size before transcription; the default sample is 4 MiB.
- ElevenLabs Scribe supplies word timestamps for accurate captions and clip boundaries.
- Upload the same audio to Zvid, then let OpenRouter select moments using transcript word positions.
- Build and validate each branded audiogram and obtain its render-credit quote.
- With dryRun=true, return editor drafts and quotes; AI and transcription charges may still apply.
- By default, render clips sequentially, retry explicit capacity rejections within a deadline, and return completed MP4s.
Setup
- Install @zvid/n8n-nodes-zvid from Settings → Community nodes before configuring credentials. A workspace owner/admin may need to install it.
- Create a Zvid API credential at https://app.zvid.io/api-keys with Base URL https://api.zvid.io. Select it on every Zvid node and Upload episode audio.
- Connect ElevenLabs Header Auth named xi-api-key, with Speech to Text permission, to Transcribe (Scribe).
- Connect OpenRouter to Pick moments. Set feedUrl, showNameOverride, handle and branding in Config.
- Test manually. dryRun=false renders and spends Zvid credits. Keep maxAudioBytes within your plan and use media you have permission to repurpose.
Requirements
- An n8n workspace supporting community nodes, Zvid render credits, ElevenLabs Speech to Text and OpenRouter access. Highlights are limited to the transcribed sample.
Customization
- Adjust colors, fonts, clip count, clip duration and sample limits in Config.
Additional info
Open Watch video → Binary → data → View and select each clip. Social publishing is not included. Multipart audio upload uses HTTP Request because the published Zvid node lacks that operation. Setup guide: https://github.com/Zvid-io/zvid-n8n/blob/master/workflows/podcast-audiogram-clips.md
🔗 Nodes Used
HTTP Request, RSS Read, Schedule Trigger
📥 Import
Download workflow.json and import into n8n:
Workflow menu → Import from File