podcast-pipeline

To repurpose podcasts into a content calendar, ingest RSS-fed audio or transcripts to generate platform-native assets like short-form scripts, threads, and LinkedIn articles. The workflow extracts editorial content atoms, applies viral scoring, deduplicates against recent history, and outputs a weekly publish schedule.

Yes, Whisper transcription is required for processing RSS-fed audio ingestion. The system uses Whisper to transcribe the audio before extracting structured content atoms and drafting platform-native social media assets like TikTok scripts and Twitter threads.

Viral scoring evaluates extracted podcast content by calculating novelty, controversy, and utility. It filters generated social media assets by a minimum score threshold to ensure only high-potential short-form video clips and posts are selected for your content calendar.

Yes, you need API access to OpenAI or Anthropic for LLM content extraction and generation. These dependencies power the editorial brain extraction, identifying narrative arcs and quotable moments, and drafting the cross-platform social media assets.

The best way to prevent duplicate content is using semantic deduplication against recent output history. Before adding newly extracted podcast quotes and threads to your weekly content calendar, the system checks them against prior assets to avoid repetition.

Yes, you can process a single episode or a batch of provided transcripts to produce a weekly content calendar. The system applies platform-specific timing rules and limits to schedule generated assets like LinkedIn articles, newsletter sections, and blog outlines.