Spotify wrapped powered by notebooklm

If you have ever wished your Spotify Wrapped NotebookLM recap felt less like a marketing graphic and more like a personal documentary about your listening year, you are about to get exactly that. I am going to show you how to combine Spotify's annual data with Google's NotebookLM so that AI generates a narrated, podcast-style breakdown of your music identity in under 15 minutes.

Direct Answer: Spotify Wrapped powered by NotebookLM is a do-it-yourself workflow where you export your Spotify listening data (top artists, tracks, genres, minutes streamed) into Google's NotebookLM, then prompt the AI to produce summaries, narratives, and even an Audio Overview podcast about your year in music. It works because NotebookLM is a source-grounded reasoning tool, so it only references the Spotify data you upload, which keeps the recap factually accurate to your actual listening behaviour.

Why This Beats the Default Spotify Wrapped

Spotify's official Wrapped is a polished slideshow, but it is also one-size-fits-all. You get the same template as 600 million other users. As someone who has trained more than 79,000 students across 74+ courses on AI workflows, I can tell you the real insight is in the layer Spotify hides: the why behind your listening patterns. NotebookLM lets you interrogate the data, ask follow-up questions, and generate a narrative that reflects your actual taste evolution, mood shifts, and discovery moments across the year.

Three things make this combination powerful:

• Source-grounded answers: NotebookLM never hallucinates outside your uploaded files, so every claim ties back to your real Spotify export.
• Audio Overview feature: Two AI hosts discuss your year in music in a 6-12 minute podcast you can actually share.
• Multi-source synthesis: You can layer in your Apple Music data, Last.fm scrobbles, or even your own journal notes for richer context.

Step 1: Export Your Spotify Listening Data

Spotify gives you two paths to your data. For a quick recap, use Stats.fm or Last.fm exports if you have been tracking already. For the deepest analysis, request your full data archive from Spotify directly:

• Go to spotify.com/account/privacy
• Scroll to "Download your data" and request the Extended streaming history (this can take up to 30 days, so plan ahead)
• For an instant recap, use the standard Account data export, which arrives in 1-3 days

You will receive a ZIP containing JSON files: StreamingHistory0.json, YourLibrary.json, and Playlist1.json among others. These are your raw materials.

Step 2: Prepare the Files for NotebookLM

NotebookLM accepts PDFs, text files, Markdown, Google Docs, and pasted text, but it does not natively read JSON. You have two options:

1. Convert JSON to readable Markdown using a free online JSON-to-Markdown converter or by asking ChatGPT or Claude to reformat the file into a clean table of artists, songs, play counts, and timestamps.
2. Generate a summary CSV sorted by top artists, top tracks, top genres, and listening time by month. Save as a Google Doc or PDF.

Aim for one master document under 500,000 words (NotebookLM's per-source limit). For a year of listening, this is plenty of room.

Step 3: Build Your NotebookLM Project

Open notebooklm.google.com, sign in with your Google account, and click "Create new notebook." Upload your prepared Spotify document as a source. NotebookLM will index it in roughly 30-60 seconds and generate an automatic summary you can use as a starting point.

Now add complementary sources to enrich the analysis:

• Wikipedia pages for your top 3 artists (gives NotebookLM context about genre and era)
• Your personal journal entries or a calendar export (so the AI can correlate listening with life events)
• A music criticism article on the year's biggest releases (for cultural context)

Step 4: Prompt NotebookLM for Insights

This is where most people stop short. The default "summarize my year" prompt produces generic output. Instead, use layered, specific prompts:

• "Identify the three biggest shifts in my listening behaviour across the 12 months and explain what they reveal about my taste evolution."
• "Compare my top genre in Q1 to my top genre in Q4 and write a short narrative about what changed."
• "List five artists I discovered this year that I now play regularly, and explain what they have in common."
• "Write a 200-word personality profile of me as a listener, based only on the data in my Spotify file."

Step 5: Generate the Audio Overview Podcast

Inside NotebookLM, click Audio Overview and customise the focus before generating. A useful focus instruction is: "Create a 10-minute podcast where two hosts discuss my musical year as if they were music journalists analysing a notable artist's listening diary. Highlight surprising patterns and seasonal mood shifts."

Generation takes 5-15 minutes. The output is a downloadable MP3 with two AI voices in natural conversation. This is the asset most people share — it sounds like a real podcast episode about you.

Common Pitfalls and How to Avoid Them

From running this workflow with hundreds of students, three issues come up repeatedly:

• Data too raw: Uploading the JSON file directly produces noisy output. Always pre-summarise into a clean Markdown or PDF.
• Prompts too vague: "Tell me about my year" gives a thin paragraph. Layered, specific prompts unlock the depth.
• Skipping context sources: Without supplementary documents, NotebookLM can only describe play counts. Add 2-3 context sources for genuine narrative.

Your Spotify Wrapped NotebookLM recap is now a personalised, AI-narrated artefact rather than a slideshow you swipe through and forget. The next step is simple: request your Spotify data export today so it is ready when you sit down to build the notebook.

---

Keep Learning

If this was useful, these are worth reading next:

• The Future of Business: Turn Your SOPs into AI Agents (Automate Everything)
• Create 40 social media posts using ChatGPT and Canva in less than 2 minutes
• Or go further with the AI Mastery Course — used by 79,000+ students across 150+ countries.