Podcast Ingestion Pipeline

Podcast Ingestion Pipeline An architecture diagram generated by Archify. Maintainer · Ops / dev · Architecture component Maintainer Ops / dev GitHub Node · Fetch / commit audio + transcripts · n8n Workflow GitHub Node Fetch / commit audio + transcripts Airtable Node · Track progress · n8n Workflow Airtable Node Track progress Whisper ASR Nodes · HTTP: send audio, get TXT/SRT · n8n Workflow Whisper ASR Nodes HTTP: send audio, get TXT/SRT Ollama Node · Embeddings · Llama3.1:8b · n8n Workflow Ollama Node Embeddings · Llama3.1:8b Supabase Node · Insert transcripts + embeddings · n8n Workflow Supabase Node Insert transcripts + embeddings runs pipeline log progress audio file TXT / SRT embeddings + transcript n8n Workflow Legend Backend Database External

Ingestion

  • • GitHub node fetches + commits audio and transcripts

Processing

  • • Whisper ASR over HTTP returns TXT/SRT
  • • Ollama builds embeddings with Llama3.1:8b

Storage

  • • Supabase stores transcripts and embeddings