YT YouTube2Text

YouTube2Text → Transcripts → Best AI Note Takers — 2026

Best AI Note Takers — 2026 — Transcript

by Jasper Tech · 4,545 words · 660 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00Hey, it's Jasper. And yes, I'm back with
0:02my arts and crafts. And today we're
0:04talking about this latest generation of
0:06AI pins. And these things have also been
0:08marketed as AI note-takers, second
0:10brains, life loggers, lots of things.
0:13But whatever you want to call them, I've
0:14been testing a ton of them over the past
0:16year or so, and I've gone from skeptical
0:18to convinced this is one of the most
0:20important consumer tech categories right
0:22now. And yes, I acknowledge we've heard
0:24this all before with things like the
0:26Rabbit R1 and the Humane Pin, but I
0:28swear these things are different.
0:31They're not trying to replace your
0:32phone, and more importantly, I found
0:34that some of them can actually be very
0:36useful. It is not just me that thinks
0:38there's something here. In the last few
0:39months, both Amazon and Meta have each
0:42purchased one of the varied companies
0:43that make these things. So there's a ton
0:46to unpack, and in this video we'll cover
0:47the currently available products, what
0:49they're all about, the results of my
0:50testing, and which ones I actually think
0:52are worth buying, and then we'll look at
0:54what's coming, because there's so much
0:55happening in the AI audio space. So with
0:58all that out of the way, let's dive in.
1:02>> [music]
1:03>> You can see I have a slightly ridiculous
1:05number of these things across basically
1:07all shapes and sizes, but an important
1:09thing to understand is at their core
1:10they're actually all very similar. They
1:12just record audio, transfer to your
1:14phone, then the app will convert that
1:16audio into a transcript, and then that
1:18transcript gets processed by an AI model
1:21to generate notes, action items,
1:22insights, basically anything you can
1:24imagine. So we could try to categorize
1:26these by their form factor, or we could
1:28also look at their platform support,
1:30since some of these also have desktop
1:31apps to capture meetings taken on your
1:33computer. And I will touch on those
1:34points, but I think the best way to
1:36understand these is by thinking about
1:38them across two dimensions. Are they
1:40always listening, or are they only
1:41recording when you trigger them? And are
1:44they mainly recording and summarizing,
1:46or they actively trying to interpret and
1:48guide you based on what they hear? So
1:50when I lay these out, you can see these
1:52two groups start to form. And I found
1:54from a practical useful today
1:56perspective, these here, the trigger
1:58recording summarization tools, are the
2:00ones that are actually very helpful. So,
2:02let's start there. In this group, we
2:04have options from Plaud, Pocket, Sound
2:06Core, Tic Note, HiDock, and a few other
2:09cheaper unbranded options, and they all
2:11promise basically the same thing, which
2:12is to be your AI note taker. And for
2:15context, I'm dyslexic, and I've been a
2:17tech product manager for about the past
2:1910 years. So, I intimately know the
2:21struggle of taking notes and the value
2:23of having good notes. So, to test these,
2:26I did bring them with me to some
2:27meetings with consent, but I also played
2:29back podcasts to them, and I used them
2:31with friends in noisy environments like
2:33bars. And after all that testing, the
2:35main thing I noticed was that the raw
2:37transcript and speaker detection was
2:39surprisingly similar across basically
2:41all of them, which got me wondering, how
2:43is that possible? So, after some
2:45investigation, it turns out they're all
2:47outsourcing their transcription and
2:48summarization needs to the same few
2:50companies. But even though they all have
2:53the same bones, I found there are five
2:55really important areas where they do
2:57differ.
2:58So, first and probably most important is
3:00file transfers. So, how easy is it for
3:02me to get the audio files off the device
3:04onto my phone? Because without that,
3:06nothing else really matters, and
3:07surprisingly, some of these are actually
3:09pretty darn bad at that. Second is the
3:12app quality and ecosystem. So, how good
3:14is their mobile app, and do they offer a
3:16desktop app for meeting recording? Third
3:18is trust. How much do I trust the
3:20manufacturer with my recordings, and how
3:22much confidence do I have that they'll
3:24continue supporting the product into the
3:25future, since I don't want a really
3:28expensive paperweight. Fourth is data
3:30lock-in. So, how easy is it for me to
3:32get my notes out of their app and into
3:34other systems? And finally is cost, of
3:36course, both upfront and ongoing costs.
3:39So, [snorts] after taking their
3:40performance and all these factors into
3:42consideration, I've got my winner and
3:44which ones to watch. And I won't string
3:46you along. The one I'm going to keep
3:47using is Plaud. And to confirm, this is
3:50not a sponsored video. I paid for these
3:52myself, and I actually might upset Plaud
3:54in a second. So, be sure to stick around
3:56for that. But, here's how I got to Plaud
3:58being my winner. In my testing, Plaud
4:00was by far the best at file transfers
4:02from both a speed and reliability
4:04perspective. The transfers can happen
4:06automatically in the background over
4:08Bluetooth, which I found worked really
4:09well. And if I needed to transfer a
4:11longer recording, I could have the
4:13device spin up a Wi-Fi network and then
4:15connect my phone directly to it for
4:17faster transfers. Now, many of the other
4:19devices also offer similar
4:20functionality, but I found none were as
4:22reliable or as fast as Plaud. Plus, if
4:25you have the Pin or the Pin S, you can
4:28have them automatically connect to your
4:29Wi-Fi network when they charge, and then
4:31they can transfer the files directly to
4:33the cloud. And from a software and
4:35ecosystem perspective, the Plaud app is
4:37not the most beautiful from, like, any
4:39stretch of the imagination, but I found
4:41it does work well. It has a good AI
4:43search where I can ask questions about
4:44my notes, and it has a solid selection
4:46of different AI models that are used to
4:48create the summaries, plus a good
4:50template library, which are basically
4:52like the pre-built instructions that
4:53shape the format and style of your
4:55notes. I also tested their desktop app,
4:58and that worked well, allowing me to
4:59automatically get notes from Zoom
5:01meetings and Google Meets and whatever
5:02tool I happen to be using. And since it
5:04uses the system audio and mic, there's
5:07no bot in the meeting, and this lets me
5:08also use my own headphones. And in
5:10general, I just really value having the
5:12companion desktop app because it allows
5:14me to have my in-person meetings and my
5:16virtual meetings and all the notes in
5:18one place. But now, let's talk trust
5:21because this is a bit more of a tricky
5:23one. When you use these, you're letting
5:25another company handle your personal
5:27recordings and transcripts. So, I'm not
5:28going to just let any random company
5:30have those, but it's tricky because from
5:32the outside, it's really hard for me to
5:33know which of these are actually
5:35trustworthy. But, I did do some due
5:37diligence, and I can say that Plaud does
5:39at least seem to be making the right
5:41noises. They claim to encrypt and
5:43anonymize the recordings, and they have
5:45SOC 2 and HIPAA certifications, which
5:47supposedly means even a doctor could use
5:49it. I'm not saying this means they're
5:50perfect or that you're never going to
5:52have any problems, but they do seem to
5:54be a lot more legit than many of the
5:56other players in this space. I'll also
5:58say that Pocket also looks to do quite
6:00well here, but I'll talk more about them
6:02in a bit. And then the other area of
6:04trust is do I think the manufacturer
6:06will continue to exist or even support
6:07the product into the future? Because all
6:09of these rely on a server connection,
6:11and there are just so many examples of
6:13perfectly good hardware going to waste.
6:15And again, I can't guarantee that Plaud
6:17won't be an issue, but of all the
6:19players, they've been doing this since
6:202023, and they're by far the most
6:22established. But now, let's talk about
6:24another area that's near and dear to my
6:26heart, and that is data lock-in. What I
6:28mean by this is if you frequently use
6:30it, you'll quickly build up a large
6:31collection of notes. And that's great,
6:33but it's also where many companies will
6:35trap you. There's actually a great book
6:37called Seven Powers that allow mental
6:38companies can build a durable
6:40competitive advantage, and one way is to
6:42build products with high switching
6:44costs, which basically means that once
6:45you're in, they're a pain to get out of,
6:47and I think these count. So, under that
6:50lens, the two options that looks the
6:51best from my perspective are Pocket and
6:54Plaud. Pocket offers an MCP server,
6:56which I won't get into here, but it's
6:58great for people who know how to code,
7:00and Plaud has three good things going
7:01for it. First is a Zapier or Zapier
7:04integration, however you pronounce it,
7:06which lets you automatically send your
7:07notes into basically any system you
7:08want. Second is there's a setting to
7:10automatically email you your notes, and
7:13third is there's a great developer
7:14community building some really cool
7:15stuff, but I'll talk about that in the
7:17next section. So, finally, let's talk
7:19about the costs, and here's how much the
7:21hardware for all of these runs as of
7:23recording, and keep in mind these may
7:25change. And honestly, this entire
7:27category is changing pretty quickly, so
7:30even though this video is published, I'm
7:31going to keep on testing. So, for the
7:33most up-to-date information, be sure to
7:35check out my site Dip. There's a link in
7:36the description. On there, you'll find
7:38my latest thoughts along with any
7:39changes to my recommendations. There's
7:41also a quiz that I'll keep updated to
7:43help you find the best option for your
7:45specific needs, and you can sign up for
7:47deal alerts on any of these, so you can
7:48snag a good deal when their price dips.
7:51And for transparency, using my links
7:53does help support the channel at no cost
7:55to you, but enough of all this, let's
7:57get back to the video. An important
7:58thing to keep in mind is there isn't
7:59ongoing costs associated with using
8:01these. Every time you transcribe or
8:03summarize a meeting, someone, somewhere
8:05is getting a bill because it does
8:07actually cost a surprisingly large
8:08amount of money to power the servers
8:10that run these things. And this is where
8:11things get interesting. Some companies
8:13will offer cheap hardware, but then
8:15they'll get you on an expensive
8:16subscription. Then other companies will
8:18lure you in with generous limits, that's
8:20often not sustainable.
8:22So, here's a quick overview of how much
8:23all their subscriptions cost, and again,
8:25this is as of recording, so things may
8:26change. And you'll see that Plaud is
8:29kind of on the expensive side, but if
8:31you're willing to be a bit technical,
8:32there's actually a way to make it much
8:33cheaper, and Plaud will probably hate me
8:35for talking about this, but my adoring
8:37fans, you guys need to know, there's
8:38actually an independent project called
8:40Audio Bridge that's you bring your own
8:42API keys to power the transcription and
8:44summarization. This makes things much,
8:46much cheaper. Now, there's no guarantees
8:48how long this will work, but I will put
8:50a link to the project below. The three
8:52others that I think might be worth a
8:53look are the Soundcore Work, Pocket, and
8:56HiDock. So, first let's talk about the
8:58Soundcore. Overall, I like the hardware.
9:00It has this carrying case that also
9:02charges it, plus there's a built-in
9:03magnetic clip, but I think it's really
9:05the app that lets it down. It's actually
9:07the same app that you use to configure
9:09your Soundcore headphones, and that just
9:10feels kind of off to me. Plus, there's
9:12no AI search, you can't pick what model
9:14is used to summarize your notes, and I
9:16can't get their Notion integration to
9:18work. And in general, it just feels like
9:20a V1 product made by a company that
9:21knows how to build good hardware, but
9:23that isn't the best with software. That
9:25being said, you know, Soundcore, they
9:27are a respected brand, and they're known
9:29for having good support. So, they're
9:30kind of one to watch, and hopefully they
9:32can get the app in a better place soon.
9:34Next, we have Pocket, and this one's
9:36tough because I want to like them. They
9:38have nice hardware that's metal with
9:40that USB-C port and built-in magnet, so
9:42there's no case needed. The phone app
9:43feels refined and they're working on a
9:45desktop app. Plus, they have a young
9:47ambitious team that's in Y Combinator.
9:49And honestly, I just want to support the
9:50underdog, but I've got two hesitations.
9:53First, I've had a decent number of
9:54issues with being able to reliably
9:56record and transfer files to my phone.
9:58They have put out some firmware updates,
9:59which have helped, but it still hasn't
10:01been as reliable as Plaud. And second,
10:04they recently flip-flopped surrounding
10:05what their free plan includes. They went
10:07from unlimited everything with 90 days
10:09of access, which is just super generous,
10:11to then suddenly changing it so it's
10:12only 14 days of access. And then their
10:14users, you know, rightfully revolted.
10:16So, they undid it and went back to 90.
10:18And I do give them credit for listening
10:19to their users, but in general, it just
10:21feels like they have a lot to figure out
10:22both with regards to their product and
10:24their business. So, unfortunately,
10:26they're one I can't recommend right now,
10:28but hopefully once the product matures,
10:30they become a real contender.
10:32The last one I want to talk about in
10:33this category is the Haydock P1. It's
10:36conceptually very similar to all the
10:37other options we just discussed, but it
10:39is unique because it allows you to pair
10:41your Bluetooth headphones directly with
10:42it, and then you plug it into your
10:44computer. This allows it to invisibly
10:46record and summarize conversations
10:47without needing to install any software
10:49on your computer because it's saving the
10:50files locally on its own memory, and
10:52then you use a web app to access it. And
10:54you can also use it as a normal room
10:56recorder, too, if you want to. And they
10:58have the P1 Mini, which is the same
10:59idea, but it's designed to work with
11:01your phone instead of your computer. And
11:03based on my testing, I can say they are
11:04bare-bones, but they do work well and
11:06basically as you would expect.
11:08Personally, if I wanted to use an AI
11:09note-taker with my own headphones on my
11:11computer, I would just use a
11:12software-based solution, but I get there
11:14are some situations where maybe your
11:16computer's like really locked down and
11:17you can't install any software on it.
11:19So, I can see how these would make sense
11:20for those people. Next, let's talk about
11:22these. And these are the ones that
11:24listen all the time to collect as much
11:26context as possible about your life, and
11:28they feed that into their AI systems.
11:30And the high-level pitch here is that
11:31they become this all-knowing second
11:33brain with perfect recall, proactive
11:35help. You know, the sky's really the
11:37limit, but unsurprisingly, there are
11:39some issues. First, let's get Friend and
11:41Lucky out of the way, and I won't
11:43sugarcoat it, they're just terrible.
11:45Friend is sassy and genuinely the
11:47equivalent of having an annoying person
11:49eavesdropping on you and then texting
11:50you about it, so I would not recommend.
11:53And Lucky is interesting because in
11:54addition to the microphone, it also has
11:56a camera. But first, it just doesn't
11:57work well, and even if it did, it looks
12:00like you're wearing a police body cam.
12:01Like people were absolutely staring at
12:02me when I wore it around, and that
12:04immediately makes it a non-starter.
12:06Also, I'm not going to waste your time
12:07talking about the Limitless Pendant
12:09because when Meta bought the company,
12:10they took that product off the market,
12:12which is frustrating because it was one
12:13of the stronger options, but you can't
12:15buy it, so it's gone. And that leaves us
12:17with Omi and Bee and Fielday. And at a
12:20glance, these three are quite similar.
12:22They all offer conversation overviews
12:23that summarize what you've talked about,
12:25an AI-powered search to help you recall
12:27anything from those conversations, and
12:29they do daily recaps, so it's just kind
12:30of like having your life narrated. I'm
12:32not sure how useful this is, but it is
12:33sometimes fun, and auto-generated
12:36to-dos. So, the real question in my eyes
12:38isn't what they do, but how well they do
12:40it, and the answer to that is
12:42complicated. So, starting with Omi, I
12:45want to like them. They're open source
12:46and they're pushing boundaries, but man,
12:48that team is just biting off way too
12:50much. Want a desktop app that can record
12:53your screen so everything you see is
12:54also part of your memory? They've got
12:56that. Want AI glasses? Well, they've got
12:59that. Want something that can read your
13:00mind? Well, they're working on that,
13:02too. They're just working on so much,
13:04and I do respect the ambition, but based
13:06on my experience, everything is just
13:07kind of buggy and doesn't work reliably.
13:10And it's not an isolated hardware issue
13:11because look, I actually have three of
13:13these things. So, in my opinion, they're
13:15just a classic example of peanut
13:16buttering, where they're spreading their
13:18team way too thin across way too many
13:20areas, and instead, they need to focus
13:22on getting their core features working
13:23properly. So, until this is improved,
13:25for me, it's sadly relegated to being a
13:27toy. Next, we have B, and I got mine a
13:30little over a year ago, and initially I
13:32was pretty impressed. It was first to
13:34market, and the app felt polished, and
13:35unlike Omi, it was actually pretty
13:37reliable. Now, this doesn't mean it
13:38didn't have any issues. Believe me, it
13:40did, but it did at least work
13:42consistently. Sadly, ever since Amazon
13:44bought them, things have gone downhill.
13:46The USB-C port of mine died after very
13:48little use, and their customer support
13:50was just totally silent. And even after
13:52spending my own money to get a
13:53replacement so I could keep on testing,
13:55the app has been buggy with my
13:56recordings randomly getting stuck in
13:58processing. I do suspect Amazon is
14:00working behind the scenes and updated
14:01hardware and software, and they kind of
14:03put things on ice while they focus on
14:04that, but it's lame to see things pan
14:06out this way. Also, based off what I've
14:08seen Amazon do with Alexa, where they
14:10started using it to advertise products,
14:12I worry about what they'll do with B.
14:14And finally, there's Fielded. And I said
14:16these all have similar core features,
14:17and that's true, but B and Omi both have
14:19some flourishes. For example, B has a
14:21feature to help you introspect, and Omi
14:23has an app store of integrations.
14:25Fielded doesn't really do any of that.
14:27It's just got the core with
14:28conversations, tasks, and search. And I
14:30think that's actually a strength because
14:32they're not overextending, and from my
14:33experience, of all these options,
14:36Fielded is the one that works the most
14:37reliably and produces the cleanest
14:39transcripts and summaries. They also
14:41have a solid desktop app, the battery
14:42lasts multiple days, and more refined
14:44hardware is coming soon. Editor Jasper
14:47here, and would you look at that? The
14:48hardware did come in the mail in the
14:50nick of time, and it is looking very
14:52polished. Also, I want to set the record
14:54straight that Fielded does offer a
14:56select few integrations. So, of this
14:58group, I think Fielded is probably the
15:00best, but here's the thing. They're
15:02actually all kind of tragically flawed
15:04if your goal is to use them as that
15:06magical second brain. And that's not
15:08just because of the massive ethical and
15:10legal issues these things face, which I
15:11promise you, we will talk about. It's
15:13because they all struggle with figuring
15:15out who said what. So, for example, they
15:17don't know if I just bought a Ferrari,
15:19or if it was a person I'm watching on
15:20Instagram, or my friend, or the random
15:23guy talking loud next to me. So, then
15:25when they attempt to proactively make it
15:26to do for me, well, now I'm being told
15:28to buy insurance for a Ferrari that I
15:30sadly don't own. This actually brings us
15:32to the bigger issue with these, which is
15:34context raw. And what I mean by this is
15:36when you indiscriminately collect all of
15:38this audio, some of it might be right,
15:41but a lot of it will inevitably be
15:42wrong. Wrong names, people, words, you
15:45get it. And now the burden is on you,
15:47the user, to filter and correct it.
15:49Because if you don't, and you just put
15:51raw garbage into the AI system, you're
15:53going to get garbage out. And it turns
15:55out that taking the time to correct
15:56everything is actually a pretty big
15:57burden. And in my eyes, these issues
16:00make the entire promise of an AI-powered
16:02second brain just not viable with the
16:04current tech. Sorry for the
16:06interruption, just one very quick note.
16:08I'm trying to get to 100,000
16:09subscribers. So, if you like what I'm
16:11doing, you like the quality of my
16:12videos, and you want to see more, the
16:13best way to support me is by subscribing
16:15to the channel, and I promise I won't
16:16let you guys down. But, let's get back
16:18to the video. All right. So, I couldn't
16:21make a video about these without
16:22acknowledging how problematic they can
16:24be. And I'm not a lawyer, I'm just a
16:26random guy on YouTube, but from a legal
16:28perspective, it does seem like around
16:3040% of the US population live in a state
16:33that requires all parties to consent
16:35before being recorded in private. And in
16:37Europe, there are so many laws that
16:39impact these products. But, those are
16:41just, you know, like the legal issues,
16:42because there's also the human side of
16:43things. For me, it's just not cool to
16:45record people in private without their
16:47consent. And unless you're in like a
16:48professional context where note-taking
16:50is expected, the act of getting consent
16:52can be super awkward because it's just
16:53not normal to record people. And then
16:55when you do, it often makes them
16:56uncomfortable and causes them to act
16:58differently. But, you know, I'm not here
17:00to lecture you or virtue signal, I just
17:01had to get that off my chest. So, now
17:03let's talk about what's coming, because
17:04things are about to get wild. So, first,
17:07there's an explosion of plot
17:08competitors, which isn't really
17:10surprising because at their core, they
17:11use commoditized hardware and readily
17:13available AI models. But, do keep in
17:15mind my five evaluation points from
17:17earlier, because sure some may win on
17:19price but there's often going to be a
17:20gotcha. In second there's a bunch of
17:22ring based options coming from Sandbar,
17:24Pebble, Fable and I'm sure many others
17:26and I'm excited to test these so Sandbar
17:28team I know you're also in New York and
17:30if you're watching this hit me up cuz
17:31I'd love to play with it. And to Pebble
17:33the Index one looks cool but the fact
17:35you can't recharge it just makes me sad.
17:37Another category I'm seeing crop up are
17:39ones geared towards self-improvement. I
17:41mentioned earlier that Bee added a daily
17:43insights feature which they claim hopes
17:45to surface patterns you never catch
17:46yourself but I'm also seeing dedicated
17:48products for this like Nerva which is
17:50designed to help people grow through
17:52deep self-awareness and Nuna an emotion
17:54tracking pendant and I'm sure there will
17:56be many more.
17:58And then we have products that go beyond
17:59audio and add cameras and yes the most
18:01obvious place to integrate this tech is
18:03in smart glasses so what do we think?
18:06How do I look?
18:07Anyway, Meta recently acquired Limitless
18:10and of course they have their Ray-Ban
18:11display glasses so it feels like it's
18:13just a matter of time before they
18:14integrate that tech but there's so many
18:16more coming like Pixel and Omi and Rokid
18:18and even though I think the body cam
18:20form factor is destined for failure even
18:22Motorola teased an AI pendant with a
18:24camera on it at CES this year. And
18:27finally we got to talk about the big
18:29dogs Open AI of ChatGPT is rumored to be
18:32working on an AI audio device after they
18:34bought Jony Ive's company. Ayo is
18:36another company making a device and
18:38they're actually suing Open AI and then
18:40of course there's Apple. They're also
18:42rumored to be working on an AI pendant
18:44and they did just buy QA I for more than
18:471.5 billion dollars and that's actually
18:49their second largest acquisition ever
18:51and that's for a startup that supposedly
18:53makes tech to decipher speech without
18:55sound by reading micro movements on the
18:57face and jaw. So we'll see what they do
18:59with that and I haven't heard any rumors
19:01about Google but you know it's just a
19:03matter of time because you know what
19:05advertising companies love? Well, that's
19:07personal data. So with all this hurtling
19:10towards us I've been thinking a lot
19:11about how this plays out. Will AI audio
19:13devices go mainstream or will they go
19:15the way of Google Glass? And for those
19:17who aren't familiar, Google released
19:19smart glasses way back in 2013, but they
19:21looked goofy and had a camera and the
19:23wearers were universally declared as
19:25glassholes and this ultimately caused
19:27the project to be canned. But I think
19:28this will be different. First, because
19:30unlike Google Glass, there's a limited
19:32social enforcement mechanism here
19:34because at least for the audio-only
19:35devices, the microphones can basically
19:37be invisible. And second, I actually
19:39think there's an interesting parallel
19:41here with professional sports. When
19:42Steph Curry started hitting all
19:43three-pointers at a rate no one had seen
19:45before, teams had a choice. They could
19:47either adapt or lose. Within a few
19:49years, every team in the NBA was
19:51building around that three-point shot.
19:53Because if you didn't, you got left
19:55behind. And I think this tech could
19:56follow a similar path. If it gets good
19:58enough that users actually gain an
20:00advantage, then regardless of social
20:02acceptability and because it's hard to
20:04detect, people will use it. And once
20:06enough people do, you'll be at a
20:08disadvantage if you don't. Now, I admit
20:10there's a lot that needs to happen
20:12before we get there, but this tech is
20:13moving at a rapid clip, so who knows?
20:16So with all that, let me know what you
20:17think. Drop a comment and if you like
20:19this video, I'd love it if you'd like
20:20and subscribe. This was a ton of work to
20:22research, test, write, film, edit and
20:24I'm a one-man band, so I'd really
20:26appreciate your support. But until then,
20:28I will see you on the next one.

About this transcript

This page contains the full transcript of Best AI Note Takers — 2026 by Jasper Tech, generated from the public captions YouTube serves with the video. The transcript has 4,545 words across 660 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.