Tech EDGE, AI in the Classroom: NotebookLM

Tech EDGE Author

05/09/2025 Added

2 Plays

Description

AI in the Classroom, examines Google's NotebookLM, an AI-powered research assistant. University of Nebraska-Lincoln (UNL) | College of Education and Human Sciences (CEHS) | Department of Teaching, Learning and Teacher Education (TLTE)

Searchable Transcript

Search:

[00:00:00.210]Coming up on "AI in the Classroom," we're playing
[00:00:02.970]with Notebook LM.
[00:00:09.780]Hi, my name is Guy Trainin
[00:00:11.550]and this is "AI in the Classroom."
[00:00:13.560]And today we're going to talk about Notebook LM.
[00:00:16.050]Notebook LM came out just about over a year ago
[00:00:19.350]and kind of took over the AI world in stone.
[00:00:23.250]And the reason they did was
[00:00:25.740]that this was a very practical application,
[00:00:28.800]easy to work with that produced a number
[00:00:31.530]of really neat results if you want to learn something new
[00:00:35.940]or if you want to process information and bring it together.
[00:00:38.910]So let me just show very quickly some of the options.
[00:00:44.160]So here I'm inside Notebook LM.
[00:00:47.910]There are multiple projects
[00:00:49.830]and I went straight into one of the projects that I have.
[00:00:53.160]This is a paper that I have written a while ago
[00:00:56.190]about computational thinking,
[00:00:58.050]and this is just an illustration of
[00:01:00.120]what you can do with Notebook LM.
[00:01:01.680]Now, in this case,
[00:01:02.700]you can see on the left hand side we have our resources.
[00:01:05.670]I have only one resource,
[00:01:07.110]but you can bring multiple texts into Notebook LM
[00:01:10.770]to process them together.
[00:01:12.450]There's a limit to how many,
[00:01:14.700]and that I think is a really actually good feature.
[00:01:18.270]So you can't just throw things in.
[00:01:20.490]You actually have to be thoughtful about what you bring in
[00:01:23.550]and what you keep out as a way
[00:01:26.250]to keep the integration from just exploding
[00:01:28.980]and becoming really unclear.
[00:01:30.930]So you can see that the first thing
[00:01:33.811]that it did in the middle is really provide a summary,
[00:01:39.870]but at the same time, on the right hand side,
[00:01:43.410]you can see what are different products that were created.
[00:01:47.460]So you can create an audio overview,
[00:01:50.490]you can create a mind map,
[00:01:52.320]you can create frequently asked questions,
[00:01:55.410]and a study guide and even a timeline.
[00:01:58.590]Now, a timeline in an academic paper is not very meaningful,
[00:02:02.280]but if it is a narrative,
[00:02:04.050]it is actually incredibly meaningful
[00:02:06.330]because if you're lost in the timeline,
[00:02:09.090]like I am usually in a "War and Peace"
[00:02:11.550]or something like this, it's just like
[00:02:13.980]too many things are happening at once.
[00:02:15.750]What is happening where, and all of that.
[00:02:17.430]These are features that can be incredibly helpful.
[00:02:19.860]It doesn't mean that you're replacing reading,
[00:02:22.830]but what it does help is organize the reading
[00:02:26.010]and make sure you notice the main things
[00:02:29.400]inside that reading.
[00:02:30.390]And it is really, really thoughtful.
[00:02:33.000]So for example, I produced a briefing document about this.
[00:02:36.510]So it has certain features including date and the subject,
[00:02:40.590]and then provide a summary and key ideas
[00:02:43.650]and themes, which is a really organized way
[00:02:47.730]to get a sense of what the paper is doing.
[00:02:50.970]The one thing it doesn't do, which I think
[00:02:55.200]that is really, really important,
[00:02:57.060]and you have to go to a different kind of AI
[00:03:00.780]or really to spend time thinking yourself,
[00:03:03.510]which is probably the best thing to do.
[00:03:05.610]And that is the critical view about this paper
[00:03:09.000]or about this product, in this case,
[00:03:10.890]in an academic paper sense, really getting a summary
[00:03:14.850]and then thinking about what isn't it doing,
[00:03:16.920]what isn't clear enough.
[00:03:18.180]And all of that could be very useful.
[00:03:20.460]It's not what is happening right now.
[00:03:23.490]So right now we took a look at this feature,
[00:03:30.090]which is the briefing document.
[00:03:33.780]Another one that I love is the mind map.
[00:03:36.630]So what you can see is the mind map
[00:03:38.730]creates this graphic representation
[00:03:41.250]that you can keep opening.
[00:03:43.200]So it gives the general topic and then the subtopics.
[00:03:47.550]But you can open any of one of these
[00:03:51.240]and it will actually start opening
[00:03:54.480]subcategories within it.
[00:03:58.350]You can keep opening, you just can see
[00:04:00.540]that this is becoming smaller.
[00:04:02.760]So you can just zoom in
[00:04:04.584]and see all the details
[00:04:08.130]and you can see that you can keep going.
[00:04:10.170]You can keep dig digging deeper into this.
[00:04:13.830]And one of my favorite features in this approach
[00:04:18.510]to a mind map is you can actually export the whole mind map
[00:04:22.050]and then use it as a picture.
[00:04:24.540]And for example, most recently we took one of our papers,
[00:04:28.500]created this mind map, edited it
[00:04:30.810]because sometimes it makes additions
[00:04:33.030]that you don't agree with.
[00:04:34.500]And after editing it, we used it as a source
[00:04:38.310]for a presentation at a conference
[00:04:41.820]allowing both the audience and us
[00:04:43.830]to see how these different relationships happen
[00:04:47.310]without using a lot of words,
[00:04:49.890]which sometimes happens in academic presentations.
[00:04:53.700]And the idea is, I have the words, I just need you
[00:04:56.910]to have a path to understand how these concepts
[00:05:00.150]relate to each other.
[00:05:01.080]So this is the mind map. Probably one of my favorite.
[00:05:04.140]The thing though that most people
[00:05:08.070]love about Notebook LM is that ability
[00:05:11.490]to create an audio that goes with everything
[00:05:16.530]and the audio is created based on an analysis of the text
[00:05:25.290]or texts that are in front of everybody.
[00:05:27.780]And then having a conversation between two podcast hosts.
[00:05:31.950]Now, it gets a lot better if you give it directions
[00:05:35.820]and not just create the podcast,
[00:05:38.790]but actually give it directions or something to focus on.
[00:05:42.030]It gets better.
[00:05:43.110]It produces something between nine and 12 minutes.
[00:05:46.380]It's a conversation between two hosts
[00:05:49.320]and it can be delightful.
[00:05:52.080]I suggest, I really, really suggest
[00:05:56.010]listening to it carefully before you are sharing.
[00:05:58.410]Sometimes it's very good,
[00:06:00.120]often it has a repetitive element to it
[00:06:03.840]and you want to make sure that
[00:06:06.180]before you share it with others,
[00:06:07.950]it is as concise as it could be.
[00:06:10.200]And also that it is selective in what it focuses on.
[00:06:15.450]And so I think that playing with it a few times
[00:06:17.820]is really good.
[00:06:18.653]If you just want to get the notion about what's in the paper
[00:06:21.030]or set of papers in front of you, it is an excellent way
[00:06:25.440]to look at the world.
[00:06:26.730]And we'll just let you get a sense of what that sounds like.
[00:06:30.750]But we've done some fascinating research
[00:06:33.720]looking at this surprising link
[00:06:35.790]between building those skills
[00:06:38.160]and well, the whole process of academic investigation.
[00:06:41.160]That's right.
[00:06:42.060]We're diving into a paper here
[00:06:43.350]that explored a fresh way actually
[00:06:45.660]to teach computational thinking.
[00:06:47.310]This was for first year engineering students.
[00:06:49.740]And so just the usual, you know, programming exercise-
[00:06:52.500](indistinct) into the clear answers.
[00:06:53.550]Right.
[00:06:54.450]They looked at whether getting these students involved
[00:06:57.090]in a process of research could build stronger
[00:06:59.160]and more adaptable computational problem solving skills,
[00:07:01.830]especially what they call ill-defined-
[00:07:04.050]Messy real world challenges.
[00:07:05.550]Exactly.
[00:07:06.960]Okay, so a few things that you can do,
[00:07:09.390]even in the original, you can speed it up.
[00:07:11.430]I like listening to things at one and a half times speed.
[00:07:14.490]It also takes a little bit of the uncanny valley,
[00:07:17.070]the weirdness of the fact
[00:07:18.480]that these are two computer generated voices,
[00:07:21.360]yet they do strange things like take time for a breath
[00:07:26.340]or pause before they're saying something,
[00:07:29.340]as if they're thinking
[00:07:30.540]that's the way that it makes it feel more human,
[00:07:33.330]but also very, very strange.
[00:07:35.100]So I like it at one and a halftime speed.
[00:07:37.500]And again, one of the important things
[00:07:39.300]that you can do is you can share that audio overview
[00:07:43.440]very quickly by creating a link
[00:07:46.440]or by downloading the sound file.
[00:07:49.470]So there are lots of ways for you to do it for yourself
[00:07:53.730]as part of learning or to share it with others
[00:07:56.940]because you think there's something interesting
[00:07:59.010]or something meaningful that you want to share.
[00:08:02.340]So Notebook LM has been with us for over a year,
[00:08:05.970]a lot of dedicated users.
[00:08:08.190]It is an incredible way to support learning
[00:08:11.700]and to digest complex texts.
[00:08:14.923]I still believe
[00:08:16.560]that reading the original is critically important,
[00:08:18.990]but this allows you also to have side by side
[00:08:22.440]kind of main highlights.
[00:08:24.810]It is not perfect.
[00:08:26.820]It makes mistakes occasionally
[00:08:29.100]and it sometimes misses points
[00:08:31.470]or overemphasizes other points.
[00:08:33.600]So you want to make sure that you create
[00:08:36.780]a balanced view of this, but Notebook LM is a great tool
[00:08:40.080]that is ready for learners.
[00:08:43.462]And I'll see you next time on "AI in the Classroom."

The screen size you are trying to search captions on is too small!

You can always jump over to MediaHub and check it out there.

Comments

0 Comments

Tech EDGE, AI in the Classroom: NotebookLM

Description

Searchable Transcript

Comments icon comment

Related Channels

Comments