Transcription
Really adds a conversational element to My Vault. I can go through and selectively choose which notes I want to have a conversation with, drop them into the source section here, and then fill out the gaps that I'm trying to add into my own note-taking with things like YouTube videos, voice notes, and articles from the internet.
Hi, my name is Callum, also known as WanderLoots, and welcome to today's video on AI and Obsidian, using Notebook LM to augment your note-taking in Obsidian. This video is part of a series I'm doing on AI and Obsidian, where I'm using different AI tools that integrate with Obsidian to augment my note-taking, to augment my second brain or my personal knowledge management system. In this video, I'm going to walk through how I use Notebook LM alongside Obsidian to boost my research, my writing, and my learning as I take notes in Obsidian.
For many people, AI is daunting to use and can feel overwhelming. That's why, before I get into more integrated Obsidian plugins that use AI, I wanted to show you how I use Notebook LM with my personal notes from Obsidian. In since there's no installation or extra steps required, Notebook LM is one of the easiest AI tools I've come across in the last 3 years, and the one that I use almost more than anything else. It makes it incredibly easy for you to upload your own sources like PDFs, markdown files from Obsidian, YouTube videos, websites, and then have a conversation with those sources. It allows you to surface insights, generate automated reports and summaries, and even can produce an AI podcast where two AIs have a conversation back and forth with one another using the notes and sources that you put into Notebook LM.
The best part? Using Notebook LM is both free and private. Privacy is a key element of why I use Obsidian in the first place, so for me, any AI tool that I'm using, that I'm reviewing and looking at integrating with my note-taking system, has to also maintain that privacy. Now you might be wondering, Notebook LM is on the Internet; it's a web UI, it's a web interface that you have to upload files to, so how is this integrating with Obsidian? Well, the key here is that Notebook LM accepts markdown files, which, which is the main way that Obsidian operates; it's how Obsidian saves all of their files. One of the main reasons I've been building my personal Knowledge Management System or second brain in Obsidian is because it lets you future-proof your note-taking; it creates markdown files which can be interpreted by any other software in the future. Notebook LM is no exception, making it incredibly easy to add markdown notes from Obsidian, which I'll get more into in a minute.
Now, markdown notes are just one reason that I use Obsidian. If you're interested in learning more about the different reasons why I use it, I actually made a video here that talks about all of my favorite features and why I got into using Obsidian in the first place, so I recommend checking that video out next. Today's video is going to be more of a practical overview of how I actually use Notebook LM and Obsidian to boost my research, writing, and general expression. If you're interested in learning more about the specific features in Notebook LM and how they can be tweaked and modified and expanded for your own use case, I recommend checking out this video here, which is my Notebook LM overview. It's actually my most popular video so far, and Google themselves found, found it helpful, so that's pretty cool. If you find this video helpful, I would love if you would please consider liking and subscribing, as your support is what enables me to continue making these videos, so I appreciate it a lot. Now let's dive into using Notebook LM and Obsidian.
Okay, so now that you have a bit of a better idea on why I'm interested in integrating Notebook LM and Obsidian, I thought I'd show you some practical use cases that I've been building out over the last few months. But first, before I do that, I just want to quickly show you the privacy policy, because one of the key goals of how I want to use artificial intelligence with Obsidian is that the data that I'm providing the AI is not used in training the AI model. I want to keep my data private; I want to keep it personal, and I want to be able to have control over the data that I'm generating and how I'm choosing to integrate it with AI. So if you see here, if you go to the Notebook LM privacy policy, you can see that it says, "We value your privacy and never use your personal data to train Notebook LM." So that's why I'm preferring to use Notebook LM with Obsidian, because if I were to drop information into ChatGPT or into Claude, depending on the settings you have toggled, you may be sending that data to the Cloud Server that the AI model is running off of, and that data might be taken and used to train their model. But by default, the Notebook LM status is that they do not use your personal data to train the Notebook LM model. But you can see here that if you choose to provide feedback, it's possible that a human reviewer may see this. So this is where you have to be really careful what type of information you're uploading, because, for example, if you're dealing with intellectual property or a patentable idea, it's possible that this might shoot yourself in the foot down the road, because it's possible that a human reviewer might be able to see the patentable idea. So I actually have a whole video that goes more in-depth on what this means and how it relates to intellectual property, so I recommend checking out that video if you're interested.
And just one quick note as well: if you're a Google Workspace or Google Workspace for Education user, your uploads, queries, and model responses are not reviewed by human reviewers and not used to train the AI models. So if you're using Workspace or Workspace for Education, you don't have to worry about the privacy at all, because Notebook LM is not going to take your data, and no one else is going to see it. And if you want to go for the paid version of Notebook LM Plus, which I talk about more in my Notebook LM update video, I believe that your information is also kept private because you're using the paid version as well. So that's just something to consider when you're using an AI tool that you want to be sure you know what's happening with your data; what's happening when you upload something. So I always like to check the Privacy terms and conditions before I start using it, and honestly, I've been pretty happy with the way that Notebook LM by Google has outlined their privacy policy, making it very clear that my information is not going to be used for training purposes, so I feel a lot more comfortable using this with my Obsidian notes.
Okay, so just as a super quick overview, I'm going to show you how you can create a new notebook here, so you can just press "New," and you can see as soon as you do so, it prompts you to add sources. So you can add sources from a bunch of different areas and a bunch of different formats. So you can see here that the type of supported files are PDF, txt, markdown, and audio, and you can connect your Google Drive through Google Docs or Google Slides; you can add website links or YouTube videos, and you can just copy and paste text. Because Obsidian operates on markdown files, this makes Notebook LM a really nice complement for augmenting your personal Knowledge Management System that you're building in Obsidian, because you can just literally drag and drop files from Obsidian, from your finder or from your desktop, and you can just add the markdown file right here, begin asking questions with your notes. Similarly, if you want to, you could export your notes to PDF, which I'll explain a little bit more how you can do that; how you can merge a bunch of different files from Obsidian into one PDF to make it easier to ask questions with Notebook LM, so I'll explain how to do that in a few minutes.
Okay, so then if you had a source, you'll see it appear on the side here. You get a chat window where you can just have a chat interface as you would commonly see with something like ChatGPT or Perplexity, and then on the right-hand side here, you have the Studio, which allows for the generation of an audio overview, which is actually an AI-generated podcast where two hosts have a conversation about your source material and your chat conversation. So that's a cool feature that, again, I'll explain in a few more minutes, but I do also have a dedicated video that goes more in-depth on how I like to use this feature. And then below that, you'll have a note section, so this is where you can generate automatic notes, or you can create notes from the chat history here to save them over time, because your chats are not saved by default in Notebook. And again, I talk about all this more in my "How to use Notebook LM" video; how to take your note-taking to the next level. So I just wanted to keep that intro as brief as possible, because today I want to show you more practical use cases of how I actually integrate Notebook LM with Obsidian, so let's get into the use cases.
Okay, so I'm going to show you a few of my favorite ways that I've been using Notebook LM with Obsidian; my practical use cases that I find help me on a day-to-day basis. I use Notebook LM almost every day, and you can see here that it's sorted by the most recently used notebooks that I've been using for quite a while. I've been slowly building them out with different sources to improve the quality and generally just augment my personal Knowledge Management System in as many ways as possible. So the first use case I'm going to show you is the research and execution of ideas; to expand on an idea or a project or a plan by bringing in specialized documents; in other words, using Notebook LM as a research tool alongside my Obsidian notes to really increase the speed and understanding that I have when I'm learning complex topics. So one of the more complicated systems that I've tried to figure out over the last couple years is the Zettelkasten system. The Zettelkasten system is a note-taking model, a system that was developed by Nicholas Luhmann, and it's one of the most productive note-taking systems of all time. And I'm not going to get too in-depth exactly what this is, because I also have another video that goes much more in-depth on what Zettelkasten is and how it works, but the gist is that you're able to take sources, you're able to take information and resources and extract the key insights that you learn from these resources, the concepts, and you're able to keep those external sources separate from your internal ideas in two different streams, so that you can always tell what was your idea and what was someone else's idea. So to build out this molecular Zettelkasten system, which is my customized version of how I like to use Zettelkasten, I have dropped in a whole bunch of my newsletters; you can see all these numbered documents on the side here, but I've also brought in a bunch of YouTube videos; I brought in my book notes on building a second brain; I brought in my book notes on how to take smart notes; I brought in markdown files from Obsidian; I brought in my own YouTube videos that I've made and other articles from around the internet; and down at the bottom here, I also have a voice note that I was having a conversation with my fiancée, Taylor, about how to build out this system and how I could fit it into what I'm trying to do with my content creation and building my business generally. So we had about an hour and a half long conversation, and I was able to drop that whole conversation into Notebook LM, and it just transcribed it. So now that I have all of these sources in here, I can go to the chat, and I can just start asking questions about it.
Okay, so for example, I'm going to ask it to, "Please summarize as briefly as possible the Zettelkasten system for a new note-taker." So you can see here that it talks about the fleeting notes, literature nodes, and permanent nodes, which are the key aspects of building out the Zettelkasten system, and you can also see, because it's bringing in my molecular Zettelkasten, it's framing the different concepts as atoms and molecules, which fits more into my, my structure that I've been building out, that fits more into how my brain works. So you can see it's, it's given a pretty solid summary here; it explains tags and folders, indexes, and you can see here that it's even bringing in specific notes. So here's my YouTube outline on topics versus tags in Obsidian, so it's referencing how I like to structure my notes using my own maps of content, and it's giving a specific reference to that note that I dropped in from Obsidian. So every one of these citations here, you can see here is one on how to take smart notes; this is a markdown file, which you can see down in the bottom here. So these are notes that I've dropped in from Obsidian. So this just really adds a conversational element to My Vault. I can go through and selectively choose which notes I want to have a conversation with, drop them into the source section here, and then fill out the gaps that I'm trying to add into my own note-taking with things like YouTube videos, voice notes, and articles from the internet. And what's nice is if I find this a good overview, I can click "Save to note," and it saves a new note on the side here. But one thing to note is that if you refresh the page, you can see that it deletes the chat conversation here, but it does keep the save note on the side here. So that's one reason why you want to save the notes from the conversation, because you'll lose them. But what you can also do is, for example, I could open this note here; I could click "Convert to source," and that would add it as a source on the side here, so that the chat interface can now have a conversation with the saved notes that I've had. Another option would be to copy this note, go over to Obsidian, create a new note, call it "Zettelkasten conversation with Notebook LM," and I can paste the notes here. So this allows me to just quickly take my conversation that I've been having and drop it into My Vault so that I can go back and find this later. And like I'll get into more later on, you'll be able to then take this type of conversation and query it using an Obsidian plugin that brings artificial intelligence into Obsidian, but I'll explain that more in a little bit. But one thing I want to note here is that what I like to do when I take a conversation from AI, whether it's ChatGPT or Perplexity or Claude or Notebook LM, I like to give it a tag that says "contains AI," and what that does is, you can see here on the side, I use the Auto Note Mover to automatically move this note into a folder in My Vault that's specifically tagged with "contains AI," that way I know exactly what in My Vault has been created by me and what has been created by artificial intelligence. That's just my personal preference that I like to do in order to keep these things separate, so I can then work through my own specific ideas without having artificial intelligence-generated content bleeding into my own ideas.
Okay, and then one last thing I just want to show you quickly here is that you can generate an audio overview conversation. So I've previously generated one, and basically what I did with this is I brought in all of my blog posts, I brought in all of my YouTube videos, all of my Obsidian notes that were all related to the concept of my molecular Zettelkasten system, and then I generated an audio AI overview that had a conversation tying together all these concepts. So then I downloaded the podcast, and I went for a walk and listened to it, and that helped me create the molecular Zettelkasten video for my YouTube channel. So it was interesting to get the perspective of an AI, to AIs having a conversation back and forth, and then take the insights from that and apply it to the way that I communicated the molecular Zettelkasten system to my YouTube subscribers. So just quickly, I want to note that this Zettelkasten video, the one that goes through molecular Zettelkasten specifically, I have a long-form video here that goes through and explains it much more in-depth. I've had a lot of positive feedback that it's really helped a ton of people, so if you're interested in learning more about that particular note-taking system, I recommend checking out that video. Just a quick reminder that if you find this video helpful, I would love if you would please consider liking, subscribing, and sharing with a friend. AI can feel overwhelming for a lot of people, so I'm trying to make AI knowledge and learning more understandable, more accessible for the average person, so I'm doing my best here to continue making high-quality videos, and your support by subscribing and sharing with other people really helps boost my ability to continue doing this in the future and make this the full-time career that I'm trying to. So I appreciate it a lot. If you're interested in getting access to my Obsidian templates or you just want to support me for the work, the time, and energy that I'm putting into making these videos, please consider joining my paid membership on YouTube. My paying members are who have been enabling me to continue making these videos, so I appreciate your support very much. Thank you. Now let's keep going with Notebook LM and Obsidian.
Now let's go to, to my next favorite feature. So another thing that I like to do here is use Notebook LM to identify themes and connections between notes and my published works. So I have one notebook called "The Plan," and "The Plan" is basically my long-term five-year plan for using note-taking systems like Obsidian, building out my digital garden, building out my personal Knowledge Management System, and then augmenting it with artificial intelligence. So to create this plan, I dropped in a whole bunch of YouTube videos; I dropped in websites; I dropped in markdown files from Obsidian; I, I dropped in voice notes, voice memos, and I took all of these 34 sources, which is many thousands and thousands of words, and now I'm able to just easily have a conversation about my entire plan. So one of the notes that I dropped in is my five-year plan and how I can bring artificial intelligence into Obsidian and then build out my website and build out my digital identity using all of this. So now I can just take all of those concepts, expand the depth of them by bringing in other technical articles, and like I mentioned before, I'm able to go just generate an audio overview. So I can take in all of these pieces of information, all of the technical documents, my research and my Obsidian notes, and I can generate an audio overview. So this just helps me make connections between different concepts that I've been working on by dropping in my plans into Notebook LM and then asking questions about how I can simplify and how I can build out my project management.
Okay, the next way that I use Notebook LM and Obsidian is to repurpose my existing content into social media or promotional purposes or use it to generate a blog outline or long-form blog post. So here you can see I have a "Social Media Generator," and within this "Social Media Generator" notebook, I have a whole bunch of different strategies; for example, ADA, Pinterest, SEO, copywriting frameworks. So I did a bunch of research on the internet and with YouTube and by conversations with Taylor, because she is a marketing expert, and I brought in all of these resources into this notebook here along with notes from Obsidian, like on different courses that I've taken and my own blog posts, and I effectively created a social media generator notebook. And what you can see here on the side is that I had a conversation with Notebook LM, and I asked it to generate my style, to analyze me, and then I worked back and forth with it to tweak my own style so that it understood what I was trying to accomplish when converting my long-form content into short-form promotional material or for expanding on my blog post. So I actually have a whole video that went much more in-depth on how I built out this notebook called "Repurposing your content with Notebook LM," so I recommend checking out that video if you want to see more how the notebook is structured and how I actually worked through to finesse out my particular style that Notebook LM now knows to use when creating the short-form content or boosting my long-form content. But for now, because this video is more on the practical side of things rather than the how-to side of things, I thought that I would show you quickly how I would actually use this notebook for one of my most recent YouTube videos. So for example, if I go over to my latest video, it's called "Master learning by teaching: The Feynman Technique using Obsidian," and basically what I did when making this video is I created a new alloy, which is the top tier of my molecular Zettelkasten system, and I went through and found a bunch of atoms related to the concept of the Feynman Technique, a bunch of singular concepts, and I built out these concepts by making analogies to different types of mental models like the hero's journey or the uncertainty matrix or metacognition or code. So I took each of these singular concepts and then I threaded them together and made comparisons to the concept of the Feynman Technique, which is effectively that you learn better when you teach people because you're able to identify knowledge gaps and then fill in those knowledge gaps, which improves your own understanding. And that gave me a lot of ideas on my own original thinking, or molecules, or depending on how you like to frame it. So I took these molecules and then, using the embed function in Obsidian, I'm able to just embed the note itself and effectively take all of my previous work and convert it into an outline of a blog post. So I took all of these molecules, all of the writing that I put here, and I sequenced the notes together so that I effectively have 80% of a blog post on the Feynman Technique done. So I thought it would be useful to show you how I can now take this outline that I have so far, add it to Notebook LM alongside the video that I used to explain this, and generate some ideas from Notebook LM that would explain how I can complete that newsletter and perhaps identify some themes or connections that I hadn't noticed when I was doing it on my own.
So the first thing I'm going to do is add a new source, add the YouTube video of the Feynman Technique; you can see it's learning; yep, there we go. So it, it has already dropped the YouTube video in, and effectively it has already transcribed the entire YouTube video. So this is a cool way now for me to take my note that I've already created in Obsidian here, leveraging the Feynman Technique, and then compare and contrast it with the YouTube video that I made where I was explaining everything I was doing to see how Notebook LM would suggest I can complete this blog post. So one thing I want to note quickly is that if I go to open this note in Finder or just in your, on your desktop, and I open it as a txt file, you can see here that there are all of these exclamation marks and then the note. So what this means is that the markdown file itself is unable to recognize that this is an embedded note that is the Obsidian user interface that's showing that a note is pulled from one spot to another. And what I mean by that is, for example, if I get rid of this exclamation mark, you can see that all of this cyclical nature of learning and teaching disappeared, but if I put the exclamation mark back, it embeds that entire note into this existing note. So because I'm trying to fill in the gaps between these outlines here, I want Notebook LM to be able to understand what the words are in the embedded note. So for this, I need to actually export this note as a PDF. So I can press Command P, export to PDF, keep the file name as title, margin all the fine, cool, click export, save it to my downloads, and boom, there. Now I have a Feynman Technique note. So you can see here it kept everything; it kept all of the embedded notes; that's what these lines are here on the side. So I was able to go to the top here, click "Add source," and then choose file to upload the PDF. So now you can see that the Feynman Technique note is in here, and it has all of the text, including all of the embedded information from the different molecules that I created. So now let's go back to chat for a moment, since I now have the Feynman Technique video and the "Leveraging Feynman Technique" note, and let's see, I want you to create the outline of a blog post using the outline section of "Leveraging Feynman Technique," the PDF; want you to analyze the Feynman Technique video and compare it to this PDF file and suggest interesting ways to convert the video and PDF into a blog post. So you can see here it took all of the ideas; it's referencing this specific note, "Leveraging the Feynman Technique," and it's referencing the "Master learning by teaching" video, The Feynman Technique YouTube video that I created. So it's identifying the intro, the core idea, how to leverage it as a tool, and then now it's actually listening to me, and it's going through and giving an analysis and comparison of the video and PDF. So the PDF is a concise, high-level overview of the core concept, whereas the video offers a much more comprehensive and practical exploration. So that's exactly the point of this; I made the video while I was writing out the outline, but I gave a lot more context in the video than I did in my PDF, in my Obsidian note. So my goal with this was, because I was just free-flowing thinking with the YouTube video, I knew that I would have an asset that I could take and then plug into Notebook LM and compare it to the notes that I took on it and then just have a conversation with Notebook LM to finesse the ideas that I had to convert it into a long-form blog post that I could then send out as a newsletter. So one of the things I was looking for was something like a title. So here you can see it suggests "Feynman Technique: From core principles to practical application," which takes in the core ideas from the PDF and then expands on it using anecdotes from the video, unlocking deeper understanding. So it's giving me a lot of really good ideas here that are very aligned with the specific notes that I included. And now let's say I like this; I'm not going to click "like," because that sends the information to a human reviewer, so I want to keep this private for myself, but I could save this to note, and you can see here that it automatically updated the note title. So if I wanted to, I could convert this note to a source by just clicking on it and clicking "Convert to source," and then I'd be able to ask questions about this even if the conversation disappears. So now, since this is a social media generator, one, why don't we say something like, "I want you to generate short-form promotional social media posts that produce the best hooks to encourage people to go watch the Feynman video or read the blog post." Okay, so here are some concepts: "Struggling to truly understand complex topics? This could be your secret weapon. Link to the blog post. Ever feel like you're memorizing, not learning? Here's the four-step Feynman Technique." You even give me suggestions for Instagram, for Pinterest descriptions, and general hook strategies that are used, which are the ADA model and the PAS copywriting framework that I included here. So you can see how it's taking in all of the different concepts, all of the different sources and information that I put in on the side here, and I'm able to specifically focus Notebook LM on answering my questions about something new, like a new video or a new blog post that I want to create.
Provides a more custom interaction with something like Notebook LM as compared to Perplexity or ChatGPT. And also, the default here is that all of these notes are private, so you don't have to worry about your information being used to train AI models or be seen by a human reviewer unless you happen to click the feedback button. So this is just a really great way for you to build on to the notes that you're taking in Obsidian.
The whole point of using Obsidian is not to just take notes that you never use, but to take notes that you're trying to actually use to improve the quality of your life or the quality of your output, which could be blog writing, digital gardening, making YouTube videos, or just having conversations with people.
Another use case for using Notebook LM with Obsidian is as a data view query builder. So I went through and I took all of the best articles, best YouTube videos, and just resources I have in general on Data View, which is a plugin in Obsidian that lets you create indexes. I talk more about this in my automatic inbox and index tutorial video for Obsidian, so if you want to learn more about how Data View actually works and what you can use it for, please check out that video. But effectively, I'm able to create custom code blocks for Obsidian by dropping in all of the different websites that explain how to use Data View.
So, for example, I could say, "Create a data view table that displays the number of links each note has within each linked note in a particular Obsidian note." So effectively, take in all of the links that I have in the note and then show me the number, number of links that are connected to each of those notes. So you can see here it generated an answer. I'm not going to test it out right now because usually I have to go back and forth a couple times to iterate on it. I just wanted to very quickly show you that you can use this to create the data view table. So you can just take this, go into something in Obsidian, add the three ticks and put Data View. So then I can just take this and copy it into Obsidian and test it. If it doesn't work, I'll go back and say, "Oh, this is the error, please work with me to create something better." So I've done this a couple times to build out my, my digital garden indexes. So you can see here it's pulling from a specific fault; it's flattening the links as atoms. So I'm able to create very complex Data View queries, and I don't have to actually learn how to code, which is the best part, because I have included all of the coding information here for Notebook LM to use. So again, it takes usually a couple iterations to get it to do exactly what I'm looking for, but it will be able to do it faster than me just trying to figure out on my own, which I think is kind of the whole point of using AI.
Okay, and now for the final use case I want to show you how I can actually use Notebook LM as a sounding board for my journals. So I'm going to create a new notebook here. The goal of this notebook is for me to take all of my daily notes from Obsidian, which are some of the most popular notes that I have. I write more daily notes than almost anything. So you can see here I have my daily note folder, and I have tons and tons of notes in here. I have 342 notes from the last several years, so almost a year's worth of notes in here if I were to take a note every day. So what I thought would be cool would be to export all of these notes into one file, drop the file into Notebook LM, and then ask questions about my own thinking; to ask questions about what jumped out at me two years ago and really to start to try and identify patterns in my thinking that will help me solve my problems better in the future. This is kind of like a personal therapist, though of course, if anyone is struggling with mental health issues, I highly recommend talking to a professional. My thought with this is that I can try to solve my own blocks and start to identify patterns in my thinking a little bit faster than just sitting there and trying to solve the problem on my own; to try and augment my thinking using artificial intelligence, which is the whole theme of this Obsidian with AI series. And this actually kind of relates to the Feynman technique video because I talk about the concept of metacognition, and metacognition is effectively thinking about thinking. So you can use your awareness and thoughts and feelings to begin to regulate yourself, so you can better plan, monitor, and evaluate how your life is going and what can make it better. This improves your knowledge over time. So my thought, thought here is effectively thoughts and feelings are what go into a journal, and if I can boost my awareness of the patterns with those thoughts and feelings, I'm better able to plan and monitor how my plans are going in the future.
So to export all 342 files, you unfortunately can't do this, as far as I'm aware, with just the Obsidian export. So we're going to need to go and introduce a new plugin. So I'm going to go over to community plugins and browse and type in "export". There's a few different options here. If you want to convert this to markdown, you can export markdown as a package, but I'm going to try the Better Export PDF because that has a lot more downloads. So you can see here one of the options is to add PDF metadata from the, from the front matter, which is good; that means I keep my markdown metadata. It supports internal links within the files, so you can click between notes; that's great if I'm referencing from one note to another. And you can export multiple markdown files, either a folder or specific files, into one PDF file. Perfect, that's exactly what I'm looking for. So let's install that, enable, and go over to options. So you can customize this however you want; if you want to change how the page numbers are being developed and just generally how the output is going to look, and you can introduce custom CSS as well if you want to have it formatted in a particular way. But for now, I just want to export all of my daily notes so that I have a single searchable file. So let's try that. And just before I do that, I want to show you one more feature about this plugin. So if you see, if you go into settings here, you can actually do a custom export where you can create a table of contents equals true document property, and then you're able to list all the notes that you want. And the reason that you might want to do this is because this will then customize the order that you want the notes included in the PDF. So I could go through and do this with all my daily notes and introduce them all day by day, and I'd be able to ensure that they're all showing up in the proper format as they are in the folder. But I'm not as concerned with this at the moment because I know I'm just going to be asking questions about this using artificial intelligence, which is not necessarily going to need to go in order to do that, but I just thought that it was a useful feature in case you're trying to export specific notes in a particular order.
Okay, so you can see here it says, in order to quick export, I can just go "export folder to PDF", and I can export the entire folder contents to a PDF file. Now, because I have so many notes, I just tried this for a second, and it will take a couple hours to do that. So instead of doing that for my daily notes, I'm going to do this with my violin practice journal because I've been learning violin, and there's only a couple notes in it. So let's go "export folder to PDF". You can see it's loading the files in the background; it's going through them one by one, and now it shows me the entire note including all the formatting, separates the dates, and it includes the metadata with each one. So I can now click export. Let's call it "practice journal.pdf", save. There we go. Now you can see here I've got my PDF with all of my markdown notes from Obsidian here. It includes all the tags; it's got everything I need. So now I can go back to Notebook LM, I can click "upload a source," choose file, bring in the PDF, and you can see here it popped up as being red; it says it doesn't work properly. So I had this before with the Feynman technique one as well, and I decided to do it a couple times before it worked properly for some reason. There we go. So that one works. So you can see it comes up with a summary: "Documents of beginner's initial experiences learning the violin over seven days." Let's ask a question: "What lessons can be learned from this experience that apply to life more generally?" Ah, here you go: "Start with small manageable steps, embrace the initial difficulties and be patient, the importance of seeking guidance, consistency is key especially when facing setbacks, reflect in your progress and adjust your approach." So honestly, this is a great summary; this is exactly the same thing that I would want to apply to my daily notes. So I'm going to export my daily notes; that's going to take a while, and my plan is to drop that in here and be able to ask questions specifically about themes and patterns that I've identified across all of my note-taking over the last couple years. So hopefully that'll give me some awareness of myself so that I'm better able to solve problems moving forward.
I hope that this shows you how you're able to use Notebook LM alongside Obsidian. You're able to generate the markdown files in Obsidian, which can be easily dragged and dropped into Notebook LM, so you can ask specific questions. You can export entire folders of notes from Obsidian into a PDF and then drop that entire PDF into Notebook LM and then ask questions about it. So there's really a lot of ways that you can use your own notes that you're taking in Obsidian and mix it with sources from across the internet, and then use Notebook LM to combine all of your thoughts and ideas and analyze them so that you can be more effective with your personal knowledge management. You can augment your personal knowledge management using artificial intelligence in a private and secure way. And there you have it; those are a few of my favorite ways to use Notebook LM and Obsidian. It really is a complimentary pairing. The ability for Notebook LM to just accept markdown files from Obsidian while allowing me to drop in voice notes, YouTube videos, websites, newsletters, and other sources makes it really a powerful tool to allow me to customize the specific questions that I'm asking to an AI system. They allow me to really hone in on specific knowledge, my personal knowledge management, so that I can augment it with artificial intelligence through Notebook LM to boost my research, my writing, and my learning generally.
That said, there are a few cons of using Notebook LM, and I thought that it would be helpful now to give a brief overview of the pros and cons so that you have a better idea on where this particular tool fits in with my overarching knowledge management system as I integrated more with AI. So the pros are that it's free; you're able to use this; there's a certain limit each day, but to be honest, I've never hit it. Part of the reason I use Notebook LM is because I tend to go back and forth and use it as a tool rather than something that thinks for me. Another pro is that it accepts markdown files, which means that it integrates perfectly with Obsidian, and it's just really a simple way for me to integrate sources from multiple places. But, for example, I wouldn't be able to use a Word document because that's under Microsoft's proprietary file format, so I wouldn't be able to use OneNote, for example; I wouldn't be able to use Word or Excel. So using markdown is a great way to keep my notes in a future-proof format so that I can continue to plug it into whatever services come out with artificial intelligence. Another pro is that you're able to bring in external sources, which makes it really nice to compare your own notes with notes and research that you've done on the internet. The audio overview feature is a great way for you to be able to distill your entire conversation and all the sources that you've brought in so that you're able to go for a walk and listen to your thought process rather than sit at the computer. And finally, you can build your own style over time in each notebook. For example, the social media generator one understands what I want with social media production, what I want to do when I want to convert my long-form writing or YouTube videos into short-form promotional materials. So I've built my style over time, and because I built this once, now I can just drop in any new YouTube video I make in the future, and it'll give you suggestions that are aligned with my own style. So that's just a cool way that you can build out each notebook almost as a dedicated tool that's aligned with the goals that you have for that particular tool.
Okay, now let's take a look at the cons. So, as with any cloud-based system, you have to be concerned about privacy. I know that I mentioned and showed you at the beginning that Notebook LM doesn't use your personal data to train Notebook LM's AI model, which is, I believe, Gemini, but if you do give feedback, it's possible a human reviewer could see it. So you are uploading something in a cloud-based system; it's up to you whether or not you trust how that's going to go in the future. For myself, I don't want to put like patentable ideas in here; I'm pretty cautious about the type of stuff I'm uploading, but that's just how I generally treat artificial intelligence tools. I take everything with a grain of salt, and I want to ensure that my best ideas stay local to my computer. So the fact that Notebook LM isn't a locally operated artificial intelligence tool to me is a huge con. Another con is that you can't suggest connections within Obsidian; like you can't have Notebook LM read the knowledge graph, so it's losing a lot of context, which is one of the main values of using Obsidian. Another con is that you have to add the notes you want each time; it's not as though the AI is updating its context constantly. You have to be very selective here. You can't save the conversations natively; they just get deleted each time, so you have to save each portion of the chat into its own separate note or copy and paste the entire conversation into your own note in Obsidian and tag it with something like "contains AI". So that's just a little manual, and there's not really any customization here. The only features that you can use this for are ones that Google gives you that Notebook LM provides, for example, like the study guide or the briefing doc, the FAQ, the timeline, and of course the conversation and the audio overview. So these are all really excellent features; don't get me wrong, but it's limited in what it can do because it's trying to appeal to everyone.
Now, all that said, Notebook LM really is a great way to get going with integrating AI with your, your Obsidian vault. It allows me to boost my ideas, my writing, my research, and really increases the speed of my learning over time. It's actually kind of mind-blowing how quickly I can learn when I'm able to take all those custom sources and then drop in my planning notes from Obsidian to have a very tailored and customized learning regime set up for me. So I find Notebook LM to be a very powerful tool that I use all of the time. But now that you've seen the easiest way to use AI with Obsidian, where you don't have to install anything separate, you can just, just go on and use your Google account, I thought it would be helpful for me to start going into more actual integrated tools with Obsidian where you can take AI and integrate it to your vault in a local, private, and free way so that you're able to take some of the features that we've talked about with Notebook LM but build it into your vault to expand it so it can actually augment your entire personal knowledge management system rather than just some selected sources. This next option is truly integrated with Obsidian, but it takes a little longer to explain, which is why I'm going to talk about it in its own video. Wow, so that only took 0.3 seconds to run my entire vault; that's wildly fast. Copilot Chat works very well; I was able to have a vault Q&A, no problem whatsoever. I can ask questions about a specific note; I can ask questions broadly on my whole vault, and it, it's able to do a semantic analysis, and honestly, it worked really well; I was impressed. This next video that I've recorded is honestly my favorite that I've made so far, so I'm really excited to show it to you.
If you found this video helpful, I would love if you would please consider liking, subscribing, and sharing; sharing with a friend. Your support really means a lot to me, and it helps me continue growing so I can continue making this my full-time career. So thank you very much. If you're looking to go deeper into Notebook LM specifically, I actually have an AI learning playlist that walks through a few of my favorite features much more in depth. I walk through how you can use the AI overview feature, the AI podcast to expand the style of your learning and your research with AI, where you can take your research and go for a walk, for example, and just listen to the conversation. It's a nice way to step away from your desk and introduce a change of pace to your desk. I also walk through how you can use Notebook LM to repurpose long-form content into social media posts, and I walk through specifically a bunch of different styles and formats and ways that you can do this effectively. And finally, I have a video that helps you understand the privacy policy of Notebook LM more so you can understand better the difference between privacy and confidentiality. Stay tuned for the next video where I actually walk through how to integrate AI with Obsidian in a safe, private, and local way. Thanks for watching, and I will see you then.