Podcast to X (Twitter) pipeline with OpenAI Whisper, GPT-4o & Slack approval

Turn your favorite podcast episodes into engaging social media content automatically. This workflow fetches new episodes from an RSS feed, transcribes the audio using OpenAI Whisper, generates a concise summary using GPT-4o, and drafts a tweet. It then sends the draft to Slack for your review before posting it to X (Twitter).

Who is this for

Content creators, social media managers, and podcast enthusiasts who want to share insights without manually listening to and typing out every episode.

Key Features

• Large File Support: Includes a custom logic to download audio in chunks, ensuring stability even with long episodes (preventing timeouts).
• Human-in-the-Loop: Nothing gets posted without your approval. You can review the AI-generated draft in Slack before it goes live.
• High-Quality AI: Uses OpenAI's Whisper for accurate transcription and GPT-4o for intelligent summarization.

How it works

1. Monitor: Checks the Podcast RSS feed daily for new episodes.
2. Process: Downloads the audio (handling large files via chunking) and transcribes it.
3. Draft: AI summarizes the transcript into bullet points and formats it for X (Twitter).
4. Approve: Sends the draft to a Slack channel.
5. Publish: Once approved by you, it posts the tweet to your X account.

Requirements

• OpenAI API Key
• Slack Account & App (Bot Token)
• X (Twitter) Developer Account (OAuth2)

Setup instructions

1. RSS Feed: The template defaults to "TED Talks Daily" for demonstration. Open the [Step 1] RSS node and replace the URL with your target podcast.
2. Connect Credentials: Set up your credentials for OpenAI, Slack, and X (Twitter) in the respective nodes.
3. Slack Channel: In the [Step 12] Slack node, select the Channel ID where you want to receive the approval request.