Summarizing YouTube Videos With AI: Three Tools Put to the Test in UX Research

You know how it is: you're researching a UX topic, find a promising conference talk on YouTube—and it's 47 minutes long. Multiply that by ten videos, and your afternoon is gone.

As a UX consultant, I've been working with qualitative and quantitative research methods since 1999. Desk research is part of my daily routine. And yes, I've spent countless hours watching videos that ultimately only contained two relevant minutes.

Since Google introduced a native YouTube summary feature with Gemini 2.0 in early 2025, I wanted to know: Can AI tools really take some of the work off my hands? And if so, which ones are suitable for professional use in UX research?

This article provides you with an honest practical test of three tools: Google Gemini, NoteGPT, and WayInVideo. You'll learn how reliably they summarize YouTube videos, where their limitations lie, and how you can specifically incorporate them into your research workflow.

📌 The most important facts at a glance

• Google Gemini is best suited for quickly screening many videos—it's free and ready to use.

• NoteGPT delivers the most detailed results with timestamps and quotes—ideal for documentation.

• WayInVideo also visualizes and organizes content as a mind map – handy for presentations and synthesis.

• All three tools reliably capture the same key points – the differences lie in format and depth.

• These tools are not suitable for confidential research videos (interviews, usability tests) – keep data protection in mind!

• The greatest added value comes from combined use: screening → in-depth analysis → visualization.

• AI summarizes, but does not interpret – the “So what?” remains your job.

  
  

The three candidates: What can they do?

Before we get into the test, here’s a quick overview of the tools. All three use large language models to analyze and summarize YouTube videos – but in different ways.

Google Gemini – The integrated all-rounder

Gemini is Google's AI assistant and has been equipped with a YouTube analysis function since February 2025. You activate the YouTube extension in the settings, insert a video link, and ask Gemini for a summary. The whole thing is free and works directly in the browser or app.

Limitation: Gemini requires videos with subtitles and officially supports English, Japanese, and Korean. However, German also works in practice, as my test shows.

NoteGPT – The documentation specialist

NoteGPT is a Chrome extension and web app that specializes in detailed summaries. The tool generates not only text, but also timestamp tables, glossaries, and quotes. The basic version is free, while advanced features cost around $2 per month.

Special feature: NoteGPT integrates AI-powered note-taking features directly into your workflow – handy if you want to process insights immediately.

WayInVideo – The visualizer

WayInVideo takes a different approach: Instead of continuous text, the tool provides a timestamped summary and an interactive mind map. Each branch contains key statements with timestamps, allowing you to grasp the structure of a video at a glance. Basic use is free.

Special feature: The tool analyzes not only audio, but also on-screen text and visual elements – theoretically even for videos without subtitles.

The practical test: one video, three results

For a meaningful comparison, I needed a video that presented typical challenges: multiple speakers, complex arguments, different positions – and German as the language.

I chose a 35-minute discussion on the topic of “Smartphone bans in schools” in the “13 Questions” format. The video features six participants with different perspectives (students, former school administrators, media educators) and ends with two compromise proposals.

Perfect for testing: Do the tools capture the nuances? Do they assign statements to the right people? And how do they structure the results?

The results in comparison

What struck me

Gemini surprised me in a positive way. The summary was not only accurate, but also cleverly structured: pros, cons, compromise proposals – exactly as the discussion format was structured. The participants were named and their roles identified (“student Lena,” “former school principal Silke”). For a quick briefing, I would immediately turn to Gemini.

NoteGPT delivered the most comprehensive result: a summary by time period, a timeline table, a glossary with explanations of terms (e.g., “cyberbullying,” “media literacy”) and even “key quotes” from the video. For academic purposes or if I want to refer back to the video later, this is worth its weight in gold. For everyday use, it's almost too much.

WayInVideo produced a mind map that visually depicts the lines of argumentation. At a glance, I can see the main topics and dive into the branches. However, some context is missing—it's not as easy to tell who said what as it is with Gemini.

The most important thing: all three were correct in terms of content

The key points were identical for all tools: addiction risk, school as a safe space, media literacy, cyberbullying, the two compromise proposals at the end. This gives me confidence in their basic reliability.

Which tool for which purpose?

After my test, clear use cases emerged:

For quick screening: Google Gemini

Want to review ten conference talks on a topic and identify the three most relevant ones? Gemini is your tool. Free, fast, well-structured. I now use it regularly in my work before watching videos in their entirety.

For documentation and research: NoteGPT

Do you need quotable statements with timestamps for a report? Do you want to be able to refer back to a video later? NoteGPT provides the depth you need. The timeline table is particularly useful when stakeholders ask, “Where exactly was that said?”

For visualization and synthesis: WayInVideo

Want to combine insights from multiple videos or present them in a workshop? WayInVideo's mind map is suitable as a basis for discussion or a starting point for affinity mapping.

YouTube summaries in UX research: Specific use cases

Now it's getting practical. Here are four scenarios in which I use AI video summarizers in a UX context – and where I steer clear of them.

1. Desk research and secondary research

The scenario: You're researching a new topic – let's say “voice UI design” – and find dozens of talks on YouTube from the NN/g Conference, UXcamp, or UXPA events.

The workflow:

1. Collect relevant video URLs

2. Have Gemini create short summaries (5 minutes instead of 5 hours)

3. Identify the 2-3 videos with the most relevant content

4. Watch them in full or use NoteGPT for detailed notes

   
   

My tip: The tools work best for conference talks with a clear structure (intro, main part, conclusion). Panel discussions with a lot of back and forth can be more confusing.

2. Competitive analysis

The scenario: A competitor has introduced a new feature. You want to understand how they are positioning it and what pain points they are addressing.

The workflow:

1. Collect product presentations, webinars, or reviews of the competitor

2. Have summaries created

3. Extract statements about features, USPs, and value propositions

4. Transfer to feature matrix or competitive analysis

   
   

Note: AI summaries give you the facts, but no interpretation. You have to answer the question “What does this mean for us?” yourself.

3. Analyze user-generated content

The scenario: You want to understand how your target audience experiences a topic. YouTube has “day in the life” videos, tutorials, and testimonials.

The workflow:

1. Identify relevant videos

2. Use WayInVideo for topic structure

3. Use NoteGPT for verbatim quotes (authentic user voices!)

4. Recognize patterns across multiple videos

   
   

Caution: User-generated content is often unstructured, jumps between topics, and contains irrelevant information. The tools provide less reliable results here than with professional talks.

4. Where I do NOT use the tools

Own research videos (interviews, usability tests): The tools presented here are designed for public YouTube videos. Confidential research data does not belong on external servers.

The limitations: What the tools cannot do

I want to be honest here, because exaggerated expectations don't help anyone:

No interpretation: The tools summarize, but they don't analyze. “This statement contradicts common UX practice” – you won't get classifications like that.

Loss of context: Tone of voice, hesitation, body language, irony – all of this is lost. This is a real problem in interviews.

Quality variation: The less structured the video, the less reliable the result. A TED Talk works better than a livestream with chat history.

Language barriers: English works best. German works, but with occasional glitches. I haven't tested other languages.

No substitute for primary research: Secondary sources remain secondary sources. For real user insights, you need real users.

FAQ:

Are these tools free?

Gemini is completely free. NoteGPT and WayInVideo have free basic versions with limitations (e.g., number of videos per day). For regular use, the affordable Pro versions are worthwhile.

Do the tools also work with German videos?

Yes, all three worked with a German-language video in my test. The results were correct in terms of content, even though Gemini officially only mentions English, Japanese, and Korean.

Can I use the tools for my user interviews?

I would advise against it. These tools process content on external servers. For confidential research data, it is better to use specialized research tools.

How accurate are the summaries?

In my test, all three tools captured the key points correctly. Errors tended to occur with nuances or when multiple speakers changed quickly. I recommend verifying important statements using the timestamps in the original.

Which tool should I try first?

Start with Gemini – free, no account required, ready to use. If you find you need more depth, try NoteGPT.

Conclusion: A useful tool in your research arsenal

Summarizing YouTube videos with AI is not revolutionary, but it does make work noticeably easier. The tools have proven themselves in my everyday work for desk research, competitive analysis, and screening conference content.

My recommendation in brief:

• Gemini for getting started and quick screening

• NoteGPT when documentation and traceability are important

• WayInVideo when you want to visualize or present content

  
  

Try it out – preferably with a video whose content you already know. That way, you can assess for yourself how reliable the summary is.

And don't forget: the tools provide summaries, not insights.

They don't do the thinking for you – but they do give you more time to do it.

💌 Not enough? Then read on – in our newsletter.

Published four times a year. Sticks in your mind longer. https://www.uintent.com/de/newsletter

Do you have your own experience with AI video summarizers in a research context? I'd love to hear from you – feel free to write to me.

Further resources:

• Google AI Studio – For advanced Gemini use

• NoteGPT – Chrome extension and web app

• WayInVideo – Summaries with mind maps

As of February 2026