Loading summary
Transcript105 lines
- [00:01]
Michael Stelzner
Hey there, Mike Stelzner. Before we get started with today's podcast, picture this. While your peers are struggling with basic chat GPT prompts, you're the marketer everyone turns to for AI solutions. You're automating tasks that used to take hours, creating stunning visuals in minutes and analyzing data like a pro. This, my friends, is not a fantasy. It's exactly what's happening to marketers. It's in the AI Business Society. When you join the AI Business Society, you get monthly live training from leading experts, real world examples you can implement immediately, and a community of innovators pushing the AI boundaries. Don't let this moment pass you by. Visit socialmediaexaminer.com AI and start your AI transformation today.
- [00:53]
Jen Lehner
Welcome to the AI Explored podcast, helping you put AI to work. And now, here's your host, Michael Stelzner.
- [01:05]
Michael Stelzner
Hello, hello, hello. Thank you so much for joining me for the AI Explored podcast brought to you by Social Media Examiner. I'm your host, Michael Stelzner, and this is the podcast for marketers, creators and business owners who want to know how to put AI to work. Today I'm excited we're going to be talking about Notebook lm. We've had an episode on this before, but today's podcast episode is going to dive into practical applications on how you can use this powerful tool from Google called NotebookLM. And I'm going to be joined by Jen Lehner, who's going to unlock it all for us. If you're new to this podcast, be sure to follow this show so you don't miss any of our future content. Let's now transition over to this week's.
- [01:43]
Jen Lehner
Interview with Jen Lehner, helping you simplify your AI journey. Here is this week's expert guide.
- [01:53]
Michael Stelzner
Today, I'm very excited to be joined by Jen Lehner. If you don't know who Jen is, is, she is an AI educator and consultant who helps entrepreneurs and marketers save time and money with AI. She's founder of the Front Row AI Club, a membership that helps entrepreneurs master AI. Her podcast is the Front Row Podcast for entrepreneurs. Jen, welcome to the show. How you doing today?
- [02:16]
Jen Lehner
I'm doing great. I am so excited to be here.
- [02:18]
Michael Stelzner
I'm excited you're here. Today, Jen and I are going to explore advanced application for Google's Notebook LLM. And if you don't know what Google's Notebook LLM hang out, because we're going to talk about it, it's absolute amazing. Now, before we go there, Jen, I'd love to hear a little Bit of your story. How in the world did you get into AI?
- [02:34]
Jen Lehner
The simplest version of my story is that really my whole life I have been interested in cool technology and how it can make our lives better. And when I get excited about something, I feel compelled to show other people, you know, why I'm so excited. Like, look, you got to try this out too. I mean, that's like the really, really summarized version. But when I was a kid, I loved robots, and Lost in Space was my favorite show. And I was fascinated by AM radio. Do you remember those? Mike, you might. You might be too young.
- [03:06]
Michael Stelzner
No, I remember both those shows. I mean, I remember Lost in Space, you know, Danger, Will Robbins. I'm. Of course, we're dating ourselves here, but keep going.
- [03:13]
Jen Lehner
Yeah. And AM radio, you could actually hear someone that might be on the other side of the country or even across the ocean. My grandfather had a CB radio in his truck, and I just could play with that for hours. Walkie talk, even those house intercoms that all my friends seem to have in their house. Most of them never work, but the idea being that you could press a button and, like, talk to somebody in the other room. I just could be entertained by that stuff for hours and hours. As I got older, I was the first person in our neighborhood to get email. And when ebay and Amazon hit the scene, I was just all over that. I even met my husband. I was the first person that I know of in our circle who had met their spouse online. It was on America Online, I think, which was before even, like match.com or any of that stuff.
- [04:01]
Michael Stelzner
I remember.
- [04:01]
Jen Lehner
Yeah, that was in 1998. So I've just always been an early adopter when I had an opportunity to be. And my first career out of college was not in anything technical. I had worked my way up to be district director of a nonprofit organization called the Muscular Dystrophy Association. And I had two states, and I produced the local Jerry Lewis Labor Day telethon each year. But even then, and we weren't really, you know, using digital tools, except I remember Microsoft Publisher and Microsoft Access, I think it was. Access? Is that. Is that what it was called? Like, you could do mail merges and stuff, and I just thought that was the coolest. And that's where I really learned about marketing, because you had to, you know, wear so many hats. But when my first child was born, I just decided to be a stay at home mom. But I couldn't stop messing with the tech. So I was learning constant contact and mailchimp And Weebly and Wix and those sites that let you build websites. I mean, you know, just templated websites. So I would volunteer for my kids schools and just ask the homeroom teacher, do you want me to make you a website? You want me to send out like a classroom newsletter? And I guess unknowingly I was keeping my skills sharp and keeping up with the digital landscape even though I wasn't getting paid for it. So when my kids were full time in school, this one day I was driving through town and I noticed this pie shop that nobody was visiting, this brand new pie shop. And I just thought, well, what a shame, they're going to go out of business before they ever get a chance. So I decided to do this cash mob, you know, where people just show up all at one time to shower the business with money. And it works so well with just some social media marketing and a little bit of email that the lines were wrapped around the building all day. They ran out of product in fact. So I decided I was going to brand that. I called it Flash Cashers and we did it for about 12 or 13 more businesses. And this was just a volunteer kind of thing that I did. But what happened was I got inside these businesses and I noticed that these were all really good businesses but they were all missing like that. They all had that gap where they just didn't know how to leverage these social media and other digital tools to get the word out about their business. So that's how I started consulting and I really loved it. And I, it honestly didn't take long for me to fill my client roster and I never really had to market because it was so easy to just make a few shifts because remember this is back in the day where really all you had to do was post on Facebook one time and like everybody would, would show up. And so I didn't have to do a whole lot before these businesses saw the ROI and had success. So I never had to market, but I could not scale my business. So that's when I discovered information products and membership sites, which was a game changer for me because I could help more people, I could finally scale. And so that's been my business model for about the last 12 years. But then here comes ChatGPT and I really couldn't sleep the day that I saw ChatGPT because I'd already been using tools like Jasper and Otter AI. But when I saw ChatGPT, I knew that everything was going to change really, really quickly. So I immediately, about a year and a half ago I changed my membership to focus solely on AI. That's the front row AI club and help members use AI without overwhelm because it's coming at us like a tsunami and I just weed through it and I, I bring the useful stuff to our members. But really if I look back, the tools have changed but the core problem that I help people with really hasn't changed and that's just to help business owners get unstuck and to charge worth and to scale without burnout. But AI is certainly the most powerful thing that I've seen for this and I just feel like I'm able to help clients create dramatically more value and to justify higher prices, which really is the same transformation I've always facilitated but amplified by like a thousand. So that was a really long answer to your question.
- [08:05]
Michael Stelzner
Totally cool, totally cool. Love the story. So folks, what's great about the Social Media examiner community is Jen is someone who's been known in that community for a while and you know, this is one of our newer shows the AI Explored podcast. So Jen and I were actually going to be doing a show on a completely different topic and then of course that topic kind of disappeared. So Jen was very, very kind and gracious and came up with a brand new topic which is what we're talking about today, which is Notebook LLM. And if you've been a regular listener for a while, we recently had Lisa Monks on the show and she did a great of high level overview of Notebook lm simply described Notebook LM is kind of like a private large language AI model where you can load it up with information and it will kind of work on just the information that you load into it and it's part of the Google ecosystem. We're going to get into some really fascinating use cases. Jen's going to describe a little bit more about how it works for those of you that are new, but specifically for marketers and entrepreneurs that are listening right now, let's talk about like why should they focus on NotebookLM said another way. What's the upside if this is done?
- [09:16]
Jen Lehner
Well, first of all, accessibility, even the free version is very, very robust. It's not going anywhere. It's owned by Google. It's getting better every single day. And remember, let's all remember Google owns YouTube, so I mean it's just going to get bigger and better as we keep moving along. And the ability to add a large amount of diverse sorts of content in different formats makes NotebookLM a dream. So think YouTube links, MP3s, PDFs, text docs, PNGs, JPEGs, website links and Google Drive, Google Docs, and now web search. You have the ability to select and deselect the content that you want to access with just a simple click. And the content, like you said, stays within Notebooks and it doesn't train the model.
- [10:11]
Michael Stelzner
Okay, so if that felt a little technical, I'll try to translate that into kind of a benefit statement. If you feel like you want something specifically that can just train on a very specific data set and not have any outside leakage. For lack of better words from the Internet or any other sources like NotebookLM is the thing that can be very powerful. And what we're going to talk about today is a whole bunch of different use cases that are very, very helpful for any business. Whether you own the business or you work for a business, you're going to find today's examples extremely beneficial. These are things you're going to be able to put to work almost immediately. So let's now get into kind of like for the total beginner with NotebookLM, just kind of explain the very basics of how it works.
- [10:57]
Jen Lehner
So what I love about it is how it just walks you through the whole process. So in fact, half the time I can't remember the link to get there, so I just type in the browser bar NotebookLM, and then it takes me to the right link. So the link is not notebooklm.google.com and then it'll prompt you. It'll say get started with NotebookLM. And then the first thing that you'll do is create a notebook. There's a giant button that says create notebook. And then it's going to prompt you to add resources. And with the free plan, you can add up to 50 resources. That's a lot of resources. And if you have Google Workspace, which you may not realize this, but you automatically have access to the Pro version, which allows you to have up to 300 sources per notebook, which is bananas, honestly, and real quick.
- [11:41]
Michael Stelzner
For those that don't know what Google Workspace is, if your company has a paid account for Gmail, then you have workspace. So keep going, Jen.
- [11:50]
Jen Lehner
Okay, perfect. Like I said, you can add where it says add sources. You can add pictures in the form of JPEGs or PNGs, PDF documents, YouTube links. And what's important here is they don't have to be your YouTube links. They could be any YouTube link. Some will get kicked out if they're private or unlisted or maybe just has like some sort of bug in it. But Most of the time you can use like any YouTube link and you can also use website links. And that alone is a huge differentiator because as you probably know, all of us that have experienced, you know, have been playing around with any of the LLMs, ChatGPT, Claude, whatever. Oftentimes you put in a link and it tells you that it can't open the link. You can also add MP3s, so that's audio files, so that means you can upload, like if you're a podcaster, you can upload your podcast. You don't even have to transcribe at first. You'll also see a newish button on the top right to discover sources. So when you click that and you type in your search query, it's going to go and search the web and give you a list of results that you can then click on and add to your notebook as a source. So the minute that you add sources to the notebook, it gets to work. And in the middle of your dashboard you'll see a summary that gets written. And all of your other sources are going to be visible along the left hand side. And as you move forward in your notebook, you can easily check and uncheck the sources that you want to interact with. And then just some of the other features, I'm not going to go deep into them, but just mention that they're there. There's an incredible mind map feature. You just click on it and it will mind map all of the content that you put in there or the content that you want it to access. There's a briefing doc feature which is an amazing resource in and of itself. If you have dense documents that you just you want. Let's say you're working in ChatGPT or Claude and you're going to exceed the content usage over there. You can reduce it with a briefing document over here, then go back over there and plug it in. If you want to stay In Claude or ChatGPT, there's a note feature, so you could create a note from anything and then again add that as a source. There's a standard chat window, so right in the middle. So that's the area where you, you just type in whatever you want to say to engage with the content the same way you do in that Chat bubble and ChatGPT and Claude, there's a study guide option, which is just what it sounds like. There's an faq, so you could create an FAQ out of anything. There's a timeline feature. I haven't used that a lot, but it's really cool. It puts things in a visual timeline and then there's an audio overview. And now a way to also interact with that audio interview.
- [14:43]
Michael Stelzner
Love it. Okay, so just kind of a high level summary. Google Gemini is what powers Notebook lm. And we don't really know which version it is. My guess is it's one of the more powerful versions of Google Gemini. And if you've been tracking, Google Gemini is getting extremely powerful. And it's, it's competitive right up there with Claude and OpenAI's ChatGPT. So the large language model behind the scenes that's kind of like, you know, powering all this is Google Gemini. What I love about the concept of Notebook. And then LM means language model. The idea is that you're loading all of these documents or all these videos or audios or whatever you want into kind of a notebook, and then you can query just what's inside the notebook. And the applications are frankly endless. But we're going to get into a couple of really fascinating applications next. And really the, the one that I want to focus on first is competitive analysis. So why don't you explain why this is something that's really valuable from your perspective and then we can get into kind of how it would work.
- [15:48]
Jen Lehner
Sure. So a way that I've recently used this is I was able to load up, I think, like 50 of your podcast episodes and then ask Notebook LM to find opportunities and topics that you haven't covered, or opportunities for me to go deeper, or just a jumping off place for me to come up with my own topics for my own podcast. So you could do that in any format with someone else's blog post, with, you know, whatever. It's, it's a wonderful feature.
- [16:20]
Michael Stelzner
Okay, so what I'm hearing you say here is, and we're going to get into kind of how this works in just a second and really unravel this. But this is a unusual competitive example that you just talked about. Right. Like, we both have podcasts in the AI world and yours is called the Front Row, mine's called AI Explored. And what you did was you loaded some of my episodes, or all of my episodes, maybe even into Notebook lm, And then you asked it, did you also load yours in there too? Or did you just ask it to identify things I have not yet covered? Like how did that work? Exactly.
- [16:52]
Jen Lehner
That would be a smart thing to do. But for this preliminary thing, I, and I do have all of my podcasts loaded in another notebook, which I'm going to Give that example in a minute. But yeah, no, it was just a simple, like, broad. It was your episodes. And then the query was, I talk about AI on my podcast. I don't want to copy episodes, but I'm looking for ideas. So where do you see, given what's listed here, where do you see gaps and opportunities for me to talk about?
- [17:23]
Michael Stelzner
Did you just literally give it the feed or what? How did you load all that stuff in there?
- [17:27]
Jen Lehner
My VA does a lot of this kind of work, but we discovered a tool called Appify A P I F I. Oh, okay. That makes it easy to do that in bulk.
- [17:37]
Michael Stelzner
So what is it? Are you pulling YouTube videos in or. Okay, yep.
- [17:41]
Jen Lehner
It was YouTube videos.
- [17:43]
Michael Stelzner
In your case, a p, p if.
- [17:44]
Jen Lehner
Y1P a p I F Y.
- [17:47]
Michael Stelzner
Tell us a little bit about what that does.
- [17:49]
Jen Lehner
It does of things. I don't really understand it. I just use it for this one thing.
- [17:53]
Michael Stelzner
Okay.
- [17:54]
Jen Lehner
Which is to go and find a lot of links on, like, YouTube and get those and put them into my notebook.
- [18:01]
Michael Stelzner
Okay. So back to this competitor analysis thing. You might be putting on a conference, like we do. Right. And there might be a competitor in your space. And I would imagine you could. Hypothetically. Let's talk that through. How would that work exactly? How. How would we use NotebookLM in that particular example? You know, maybe there's a list of speakers or something like that. How would use it to do competitive analysis on the two different events?
- [18:24]
Jen Lehner
What is our goal? We're a speaker too.
- [18:26]
Michael Stelzner
No, let's say we put on an event.
- [18:27]
Jen Lehner
Oh, we're putting on the event. Yeah.
- [18:29]
Michael Stelzner
And we want to differentiate from another event. How would we use it to do.
- [18:33]
Jen Lehner
The analysis specifically with Notebook lm? I mean, I would do.
- [18:36]
Michael Stelzner
Could you just give it the link to the speaker page?
- [18:38]
Jen Lehner
You could. In that case. I just don't know how that's different from doing it with ChatGPT or anything else. You know, I like to use examples that are. So for. Here's another example. I also used your podcast, all of those that I put in there, to find out exactly what you covered on NotebookLM, to find out what has been covered and what hasn't been covered. Because I wanted to pitch ideas to you and say, look, I know you've already done NotebookLM, but I've got all this other stuff to talk about. So I plugged in some ideas and made sure there was no overlap.
- [19:13]
Michael Stelzner
Okay. I think I'm understanding where we're going with this and I'm really glad that I'm asking these clarifying questions. So, so when you are analyzing particularly content, really that's what we're talking about here. Like instead of just competitive analysis, because like you said, you could just give the link to your competitor's website and ask Claude or Chat GPT to do the exact same analysis. But where it struggles is actually detailed content analysis. Is that what I'm hearing you say? Yes, specifically with audio and video. And that's the application where you could, like, for example, another thing that you could do hypothetically, and I'm thinking with you on the fly here, is let's just say you love a certain YouTuber and you take all of their best YouTube videos that are on a specific topic. Could you ask it to reverse engineer their flow a hundred percent?
- [20:03]
Jen Lehner
A hundred percent. That's a great, that's a great example.
- [20:06]
Michael Stelzner
So what would you ask it specifically after you popped in all those videos? Let's say I want to model how they do it. Let's say I really love Sean Cannell's style of how he does his YouTube videos. Could I take his most successful YouTube videos, plop them into Notebook LM and say yes, identify all the hooks and transitions and, and help me understand how he did this? Or something along those lines?
- [20:28]
Jen Lehner
Yes. And more and more I'm finding that starting general really will surprise you with some great results, you know, for a long time. A long time. For as long as we've been at this, you know, 10 minutes with AI. But I mean, you have to have really specific complex prompts. But, but what I'm noticing is like, it really will surprise you sometimes. So in that example, I would be really general at first and just say, you know, this is one of the most successful YouTubers out there. I know that there's definitely a pattern in the way that he's constructing his videos. Find the pattern and let me know what it is.
- [21:02]
Michael Stelzner
Ah, okay. I love that.
- [21:04]
Jen Lehner
Yeah. And then it would do that for you and then you. And then you could keep going, you know, I would build from there. That's a good idea.
- [21:11]
Michael Stelzner
What if I told you while your marketing peers are still fumbling with basic AI prompts, you could become the AI expert your entire company depends on. That's exactly what's happening inside the AI business society. Our members aren't just using AI, they're mastering it. They're automating repetitive tasks, creating professional videos in mere minutes, and making data driven decisions that leave everyone on their team completely amazed. Most marketers are barely scratching the surface of what AI can do for them. While they're stuck creating mediocre content. You could be leaping so far ahead that people look at you as if you're from the future. Here's how we make that happen. First, you get monthly live training from leading AI experts who actually use these tools in real marketing situations. Secondly, there is a incredible community of innovative marketers who are pushing the AI boundaries every day that you have access to. Third, we have multiple meetups per month where you get a chance to interact with your peers and ask questions from our experts. As member Lisa Kanda said, quote, the quality of training offered is the best I've experienced and from an organization that I can trust. Here's the truth. The marketing landscape is changing very quickly. Don't get left behind. While others race ahead. Visit social mediaexaminer.com AI and become the AI Enhanced Marketer. Your company or your client. Clients need future. You will. Thank you. Visit socialmediaexaminer.com AI to learn more. Okay, so let's go into your next example here, which was you have a case study example. Share with us a little bit about the story there.
- [22:48]
Jen Lehner
Okay. This is so exciting. I just cannot even contain myself. Okay. But I'm gonna try. So first of all, we all know we need to do case studies. But case studies are really time consuming to do well. And also case studies tend to be really dry. They just tend to be like a data dump, you know. And so I have a client who is a digital strategist. She sort of has different packages where she'll build your whole digital business. She'll do just SEO, sort of everything in between, and she'll build websites. But she's really like high touch. So for her and to work with her might be, you know, for her, her whole big package, right? To build your digital thing and your brand and all that might be $30,000. So people who are interested with, in working with Sandra Scalliono, I gotta throw her name out there, then they will read a case study, but it's gotta be compelling. So what I did with Sandra was I told her, like, the messier the better. I just wanted all the bits and pieces, the stuff that's written on the back of a napkin, screenshots of nice things that her client has said about her on social media, emails between the client and herself. I wanted Sandra to do an audio, just a rough audio overview of her work with the client. What else did we put in there? Emails, back and forth. And maybe I think that's about it. There were like five or six things we put in there. Anything that we could think of. I just wanted to dump into this Google Drive folder. Okay. When we did that, I asked the audio.
- [24:25]
Michael Stelzner
Wait, wait, hold on. You said you threw it in a Google Drive folder. Does that mean you can attach a Google Drive folder directly in a notebook? Lm.
- [24:32]
Jen Lehner
Thank you. Let me back up. I didn't. I got each of those things and I put them in a notebook. Lm Notebook.
- [24:38]
Michael Stelzner
Okay, good, good. Thanks for clarifying.
- [24:40]
Jen Lehner
Yeah, thank you. Thanks for catching that.
- [24:41]
Michael Stelzner
Okay.
- [24:41]
Jen Lehner
And then I went to the audio feature and I asked the audio feature. And what this is is that it's, it's. It's two robots. It's a male voice and a female voice. It is incredible how natural they sound. It really sounds like two human beings. It's remarkable. And I asked the robots to have a conversation and I told them specifically what this was for. I said, this is for a case study that, that my client, Sandra Scano is going to use to get more clients. So at the end of listening to this, listeners should want to work with Sandra. Okay. So I checked all this, all the boxes, so that they had all the resources and hit the button. It gave me a 14 minute audio that was so good. And here's why it's so good. This is the thing. Okay. And. Oh, okay. It automatically creates a story arc and a story. Stories sell. Stories are everything. All your marketers know this like it is. It's all about the story. So you've taken this dry content and we did this also with an SEO client of hers. I mean, what could be more dry? But it somehow turned it into this really gripping story that for 14 minutes you could listen to and really be entertained. But not everybody is going to listen to a 15 minute case study. So listen. Okay, we take the case study that lives on a landing page. Right? So you, so you make it so for all people. Right. We want to have audio, visual, you know, documents, because everybody's a different kind of content consumer. But we want to make it easy for everyone. So we did several things with this to experiment. One was you can transcribe that. My favorite transcriber these days is Whisper Apple Whisper. What's it called?
- [26:34]
Michael Stelzner
Oh, isn't that by OpenAI or by somebody else?
- [26:37]
Jen Lehner
Let me click on it. Oh, it used to be at the top of my screen.
- [26:40]
Michael Stelzner
Can you not ask NotebookLM to just write it up for you?
- [26:43]
Jen Lehner
Yeah, no, you totally can. So. No, the reason is because when you transcribe it to show Speaker 1 and Speaker 2. Right. We took that we uploaded it into Heygen that has the human avatars and we created a video conversation. Now just like we had the audio conversation. That one honestly is, it's not ready for primetime yet. Like the hey gen stuff is still just too weird looking. Then we took it and we put it into several other platforms to create interactive case studies. So I went to genspark. Genspark is another LLM. This was all done with the free plan and I said take this and create an interactive case study and then publish it as like a landing page. And it was beautiful. We did it with the Canvas AI feature as well. That one came out nice. We did it with Lovable and we did it with Gamma app and they all really came out great. So I'm just listing all of those to say, you know, it's not the end of the road when you get this audio transcript. Because this audio feature, I think that's where people sort of, they begin and end with that. They're like, that's cool. But it's not like I'm going to use that audio on my podcast. Well, that's just the tip of the iceberg. This is, it's the most incredible feature because of the story arc, because the conversation is really good and then it lends itself to all this other great content in all these other formats.
- [28:10]
Michael Stelzner
So Gen Spark is Gen like generate, like, like generative AI Spark. Okay, cool. Okay, so just for the record, what I heard you say is that you took a whole bunch of different things from your customer and you threw them all. And anybody can do this, right? They can take emails, they can take private messages, screenshots, whatever. Well, you don't want to take a private message without permission, but you get the idea. You can take all this information, maybe even a little write up behind the scene of what you did. You can throw it all into NotebookLM and you can ask it to generate a audio and to generate the audio tell everybody where they need to go to do that. Is it just called Audio Overview? I'm trying to remember what it's called.
- [28:51]
Jen Lehner
Yeah, it's called Audio Overview and it's in the top right hand corner.
- [28:57]
Michael Stelzner
And then you can download that audio file and you can transcribe it. You can even transcribe it probably inside of AI Studio, to be honest with you. And then you could take that transcript and you could have some other tool create a much better looking and story for you. And you could potentially even ask Notebook LM to write the case study for you instead of speak it out. Is that fair to assume that it's capable of doing that?
- [29:21]
Jen Lehner
Oh, yeah, absolutely. And I'm glad you said that, because what we're talking about here with doing case studies, we could dump all this and take the mess and put it into any LLM, and it's going to give us something, right? It's going to make it much faster than it's ever been. But the thing here, the differentiator, is that for whatever magic sauce they're putting on these robots, I don't know what it is. And by the way, if you've tried this two months ago, it's so much better. It used to be they used to talk about a whole lot of nothing in the beginning. All this banter, back and forth, small talk. It was a little bit annoying. Not anymore. I don't know why they get to the point. And it's like every other LLM summarizes and can synthesize and it can find patterns and stuff. But when these. I'm going to say people, right? But we know they're robots. When these people talk, this guy and this girl, when they talk, they use phrases that you go, wow, that's such a cool phrase. That's such a cool analogy. That's such a cool metaphor, the way they put that. Or if they find something funny or lacking, or they get a little sarcastic, too. They can be a little snarky. It's just next level.
- [30:31]
Michael Stelzner
So I love it.
- [30:32]
Jen Lehner
Sparks are flying, you know, in your head of, like, how you can use what they're saying.
- [30:37]
Michael Stelzner
I get it. So the key thing here is to go from kind of the messy text into the spoken audio podcast story, for lack of better words, that you can make videos out of and all these other kind of things. And I do agree that when these two characters talk, they're highly animated, they're very optimistic, and it's a really great story that they can make out of almost nothing. All right, let's move on to another example. What's your next example?
- [30:59]
Jen Lehner
The next example I want to give you is another one that really just made my heart beat so fast when I did it. I didn't know if it was going to be possible, but it's possible, basically. So I have a podcast coming up. I mean, my 100th episode is coming up, and I wanted to do a compilation episode. And I've been fortunate over the years to have some big names. I've had Gary Vaynerchuk and James Altucher and Seth Godin and some of my favorite authors like Greg McCowan and who else did I put in there? Mike Mikowski. Prophet first. I know I'm saying his name wrong, but whatever. All these people. And I could get, you know, more mileage out of these big names if I do a roundup episode. So what I did was I uploaded. I have all these episodes in a notebook folder. Notebook LM.
- [31:43]
Michael Stelzner
Are we talking MP3 files or what are we talking about?
- [31:45]
Jen Lehner
Yeah, they're MP3s, and specifically Amazon S3 MP3s.
- [31:49]
Michael Stelzner
Okay, so just folks that don't know what that means. That means there are direct links to Amazon where the actual files are. Is that what I'm hearing you say? Is that correct? Okay.
- [31:56]
Jen Lehner
Yeah.
- [31:56]
Michael Stelzner
So you put a bazillion links, 99 links in there from every episode you've ever done or just from the top episodes?
- [32:02]
Jen Lehner
Well, I already had this, but I just really wanted the episodes with these people in it, so I just checked the ones.
- [32:08]
Michael Stelzner
Okay, so you added the links to all these MP3 files and then keep going. What happened next?
- [32:12]
Jen Lehner
And then I said, I am doing a Roundup episode first. I want you to look at these episodes and tell me if you see any themes that pop out for a roundup.
- [32:25]
Michael Stelzner
Okay.
- [32:25]
Jen Lehner
Right. I mean, I was doing no mental work at all. I mean, like, I'm just putting all the heavy lifting on. On the AI. And then it gave me three really compelling themes. And so I like theme number two. So I was like, let's go with that one. So what I want you to do is to create a table, and I want three columns. In column one, I want the episode number. In column two, I want the actual transcript of the quote that you're grabbing. And then in the third column, I want the timestamps of where to cut. It gave that to me, like, almost instantly. So that was easy.
- [32:59]
Michael Stelzner
Was it accurate, by the way?
- [33:00]
Jen Lehner
It was accurate. Yeah. Yeah, yeah, it was. I liked it. They were good. That's not to say that that's going to happen every time. I mean, I'm sure that, you know, especially if I had done more episodes, I'm sure I would have had to do more tweaking. So then I go over to Manus. Okay, M A N U S. And I don't use Manus a lot, but I remembered that it does stuff. Like, even before these other agents started popping up, Manus was like, one of the first agents that was, like, publicly available that, like, did stuff. I don't know how else to say it, but.
- [33:31]
Michael Stelzner
And by the way, Manus lm. And what it is, is it's kind of like an agent that can surf the web. Is that probably the easiest way to describe it? I mean, it's like you can give it a task and it will go out and do a task for you. It's basically an agent. AI agent is really what it is. Right.
- [33:43]
Jen Lehner
It does things like it will. Well, you're going to see. It edited my podcast, but it. Yeah, so it. This is. This is. Hold on to your seat. Okay, so then I said this was the part where I didn't know it would work because I had. I had tried it once a long time ago with one audio clip and asked it to edit, and it had done it, but this is a much different thing. So I said, based on this document and based on the instructions on this table, please edit. And I told it that I was doing a roundup episode, and I said, please edit these accordingly according to this document. And it did it. Okay, so here's the thing. I think because I told it, it was a roundup episode, it thought it needed to do the whole episode. So it gave me one 12 minute MP3 file with all the clips, you know, merged together, which is no problem because I can download that now. My podcast editor can just insert my commentary in between each of those and throw on the intro and throw on the outro and like, call it a day. But I could not believe. And also I gave it Amazon S3 links as well. So I gave Manus the same Amazon S3 links, but I'm telling you, it edited very precisely the exact quotes. Yeah, clips. So, you know, I just couldn't believe it. I mean, I was. I thought that was pretty exciting.
- [35:08]
Michael Stelzner
Okay, so just for the record, you went into NotebookLM and you asked NotebookLM to produce a table for you. Is that ultimately what it did? And it. It had the name of the guest, the clip, and then maybe like the link to the Amazon MP3 file. Is that kind of what NotebookLM did for you?
- [35:24]
Jen Lehner
No, I suppose I could have added that as well. That wouldn't have been a bad idea. But on my table, it just had episode number, the actual transcript of the blurb, so I could see and make sure that it was accurate. And then the timestamps.
- [35:38]
Michael Stelzner
Oh, the timestamps.
- [35:39]
Jen Lehner
Okay, yeah, the timestamps.
- [35:40]
Michael Stelzner
Oh, got it. And then you must have brought that into another document and added the links back to the MP3 files, is that correct?
- [35:46]
Jen Lehner
When I went to Manus. But in the notebook, the MP3 files were all there. They were in that notebook. And I had only the Ones checked that the episodes that I wanted it to extract from love it. Does that make sense?
- [35:58]
Michael Stelzner
Love it. Okay, now you could also do this with blog posts, post content if you wanted to. Right. You could just give it links to your blog and have it identify all this kind of stuff. But you could probably do the exact same thing with other AI models as well. What I like about this one is the fact that you're dealing with multimodal content and it's pretty powerful stuff. Okay, we got another example, another case example which is research related. So talk to me a little bit about that research example.
- [36:26]
Jen Lehner
So I take a bath every night and it's just like, you know, that's my therapy. And what I do is I create like my own podcast using these, these folks to help me move through a lot of information. So obviously, like you, I have to stay on top of all of the developments in AI. You know, I always don't, I don't always want to be sitting at my desk, you know, for hours, like reading and reading and reading. So what I'll do is combine however many resources and this is when I can really go into like a white paper or even a book if I want to, and put it all into a notebook. And then I tell the guy and the girl to have a conversation about it so that I'm up to speed. Or recently I attended a conference and I didn't know who the keynote speaker was till the night before. And I was so bummed that I hadn't read his book because I would have liked to have read his book before I saw the speaker. So I just, again, for my bathtub time, I just put it in and told the robots to talk about it. And so I really felt up to speed and that I understood the content of the book and could have an intelligent conversation the next day. And this is not the same as, again, this is not the same as ChatGPT or any of the other language models because of this ability that this feature has to create a conversation that actually sticks, right? Like they talk about it in a very human like way. So it, it makes sense and they draw connections. So you just list and you know, they've added that feature where you could dial in and you can redirect them if you need to or ask a question while they're talking and it responds. The only thing though is that when you're done, the written documentation of that conversation only shows the audio content that they had. It doesn't show your interruption or your questions. So it's one of my favorite ways to really consume a large amount of information. Because yes, I could Summarize it on ChatGPT or any of the others, but for me, and I know there's a lot of people who are audio learners and the fact that they're not robotic sounding, they don't sound like robots, you really can digest the information.
- [38:47]
Michael Stelzner
We had another example of creating training for your staff, internally training. And I would imagine this could also be for your customers if you had a membership. Tell me a little bit about that example.
- [38:59]
Jen Lehner
That's just the same sort of idea is that you could take your dry sops and the videos and everything else, and because of the wonderful ability for NotebookLM to house all different kinds of diverse content, then when it comes time to how do I process a podcast episode, your staff member can listen to the instructions when they're out on a walk. I mean, doesn't sound like the most interesting content in the world, but I mean, it's just one more way to learn something.
- [39:29]
Michael Stelzner
Something question for you on this audio feature built into Notebook lm, do they make it easy to work on your phone or do you have to download the MP3? Do you understand what I'm asking? Because yeah, I don't even know if there's an app. Is there a Notebook LM app for the phone?
- [39:45]
Jen Lehner
There is a Notebook LM app. I'm opening it up right now because it's really not. It's not bad. It's pretty good actually.
- [39:52]
Michael Stelzner
My guess is you could create the audio feature on the desktop and then listen to it on your phone later or I don't know if that's true or not.
- [39:59]
Jen Lehner
Oh yeah, you can. It shows up in that folder, it'll be. It will be in that folder and all your folders show up on your phone.
- [40:05]
Michael Stelzner
Well, that's good news because that means that you could prep this on the desktop and then if you want to listen to it while you're running or walking or in the bath or whatever, you don't have to have your laptop next to you.
- [40:15]
Jen Lehner
Yes, exactly.
- [40:16]
Michael Stelzner
Very cool. Jen. This has been a fascinating dialogue. I know we've just scratched the surface of what is possible with NotebookLM and all these different use cases. If people want to connect with you on the socials, where do you want to send them? And if they want to potentially work with you or join your membership, where should they go?
- [40:33]
Jen Lehner
Well, they could go to Jenlaner.com forward/AI explorer, because I've got some goodies for your listeners over there. And then on Social Jen underscore Laner. That's L e H N E r.
- [40:47]
Michael Stelzner
Okay, so it was Jen laner.com AI explored if they want to learn more about you and your business. And then which social platforms did you say are the ones that they should connect with you on?
- [40:58]
Jen Lehner
That was Instagram. J E N L e H N.
- [41:01]
Michael Stelzner
E R Jen, thank you so much for sharing your insights with us today.
- [41:04]
Jen Lehner
Thank you so much, Mike. I enjoyed it.
- [41:06]
Michael Stelzner
Hey, if you missed anything, we took all the notes for you over@social mediaexaminer.com a64 Be sure to follow this show on your favorite podcasting app. And if you've been a listener for a while, we would love a review and we would also love it if you would share it with your friends. Also, do check out the Social Media Marketing Podcast and the Social Media Marketing Talk show. This brings us to the end of the AI Explored Podcast. I'm your host, Michael Stelzner. I'll be back with you next week. I hope you make the best out of your day and may AI help you become more successful.
- [41:37]
Jen Lehner
The AI Explored Podcast is a production of Social Media Examiner.
- [41:45]
Michael Stelzner
Just a quick reminder before you go. If you're ready to become indispensable in the age of AI, the AI Business Society is your solution. Join now and secure your discounted membership by visiting social media examiner.com AI I can't wait to see you inside the AI Business Society.