What This Workflow Does
This workflow automates turning podcast transcripts into easy-to-read email digests. It saves time by summarizing long texts, finding main topics, and making questions for listeners. The final digest is sent by Gmail as a nice HTML email.
This solves the problem of spending many hours manually editing podcast transcripts and making listener questions. The result is a ready-to-send summary email that improves audience engagement and cuts down mistakes.
Who Should Use This Workflow
This is for podcasters or content creators who get long transcripts and want quick summaries.
It is useful for anyone wanting to automate content processing and email sharing without spending extra time writing summaries or discussion points.
Tools and Services Used
- n8n: Automates workflow steps and connects nodes.
- OpenAI GPT-3.5 / GPT-4 APIs: Summarize transcripts and extract topics and questions.
- Wikipedia Node: Researches topics with detailed explanations.
- Gmail Node: Sends formatted HTML email digests.
Workflow Inputs, Process, and Output
Inputs
- Podcast episode transcript text entered manually or triggered.
Processing Steps
- The transcript text is converted into JSON for AI processing.
- The text splits into 6000-character chunks with 1000-character overlap to handle API limits.
- OpenAI models create a summary from the chunks.
- GPT-4 extracts main topics and related questions from the summary.
- Structured output parser ensures topics and questions are in correct JSON format.
- The topics split out to research each using the Wikipedia node.
- Further explanations for topics are generated using OpenAI GPT-3.5 in a multi-turn conversation.
- All results merge and convert to HTML format for email presentation.
- The formatted digest is sent through Gmail with OAuth2 authentication.
Output
- Professional HTML email containing a summary, topic explanations, and engaging questions.
Beginner Step-by-Step: How To Use This Workflow In n8n
Importing and Setup
- Download the provided workflow file using the Download button on this page.
- Open the n8n editor where workflows are managed.
- Use the Import from File option to load the downloaded workflow file.
Configure Credentials and Inputs
- Add your OpenAI API Key credentials in n8n’s credentials section.
- Set up Gmail OAuth2 credentials for the Gmail node to send emails.
- If necessary, edit recipient email addresses, distribution list, or other IDs in relevant nodes.
- If a transcript is preset in a Code node, replace it with your actual transcript text.
Testing and Activation
- Run the workflow manually using the manualTrigger node to check it works correctly.
- Verify the summary email is received and formatted as expected.
- When ready, activate the workflow to run in production.
For users wanting control or scalability, consider self-host n8n to run workflows on a server.
Customization Ideas
- Change manual trigger to a webhook for automatic transcript uploads.
- Adjust character chunk sizes in the recursive text splitter node to handle longer or shorter texts.
- Swap GPT-3.5 with GPT-4 models or change settings like temperature for creative versus factual outputs.
- Modify email formatting or recipients in the Gmail node to fit branding needs.
- Add nodes to pull topic data from other APIs besides Wikipedia for deeper research.
Common Failures and Troubleshooting
Issue: Transcript text too long for a single OpenAI API call.
Cause: Token and character limits on API.
Fix: Make sure the recursive character text splitter node splits the transcript into chunks with overlap.
Issue: Gmail OAuth2 authentication fails.
Cause: Expired or invalid Gmail credentials.
Fix: Refresh or update Gmail OAuth2 credentials in n8n credentials manager.
Summary of Workflow Impact
✓ Saves 3-5 hours by automating podcast transcript processing.
✓ Reduces mistakes made by manual editing.
✓ Produces polished email digests with summaries, topics, and questions.
✓ Sends content automatically via Gmail.
✓ Can be customized for various podcast formats or other transcription types.