 So I'm always on the lookout for really awesome applications and today I have a really cool one for you. So basically everyone I know takes notes, right? I don't think that there's anyone out there who doesn't take notes sometimes. And if you are like me, you take notes all the time. And I'm always kind of on the lookout for either fully functional note taking taking applications where I can do all the stuff that I could possibly need to do like ever note or note snook or things like that so that I can actually have a full fledged note taking system. Or I'm looking for neat note taking applications that do things in an interesting way. So today we're going to be taking a look at one of the latter and the application we're going to look at is called speech note and it does some really cool things. I do have a few complaints about it, but we'll talk about those as we go along. So let's take a look at speech note. But before we do, if you leave a thumbs up on this video, I'd really appreciate it. It really does help the channel. So let's go ahead and look at speech note. This is a QT application. So it does look better in plasma. But that really is neither here nor there. You can use it in GNOME if you want to as well. So the basic idea here is that you can talk into your microphone and it will type out what you say. Now it's more complicated than that and it does more than just that. But that's the basic idea. So let's actually create a note here and I will show you how I did this in just a minute. But let's just see the output. Speech note is a very cool application for note taking. Now as you can see, it took what I said and put it into speech. Now if that was all that it did, that would be neat enough, right? But it will also read it back to me. So if I want it to read, I just hit read and it will read it. Speech note is a very cool application for note taking. Now the one thing that I was most impressed with was it actually sounds like a human voice. It doesn't do a fantastic job always of pronunciation. So we'll listen to that just one more time. Speech note is a very cool application for note taking. It didn't do a very good job on its own name there, but other than that, it did actually a pretty good job. So those are the two main things that it does. We'll talk about the translation thing here in just a minute, but the main thing that it does is it will listen to your microphone when you want it to. You have to press a button or you can change that setting. We'll look at the settings here in a minute. And it will take your speech and put it into text. Now if I wanted to create another note, I'll just go to the next line here. The Linux cast is the best YouTube channel online screw you DT. So with that note, you can see that it even got the punctuation basically right there. I forgot a comma, but mostly it was actually quite close to what it needed to be. Now personally, the punctuation part is the part that's lacking to me the most because if you have ever used the transcribed function on like an Android phone or an iPhone when you want to insert a period, you just say the Linux cast is the best YouTube channel online period. And instead of writing the word period, it would put a period or I could do screw you comma DT and it would, you know, figure out where the comma needed to go, but it didn't do that. Instead, it doesn't look quite right, right? So the idea here is that you talk into the microphone, it writes the note for you. That's the basic idea. Now out of the box, you get none of this stuff. You have to download some language packs and it does tell you. So I'll put up the screenshot of what it tells you at the beginning. It tells you everything you need to do to get this set up right the first time you open up the application. But basically the idea here is you go up to languages, you find the language that you want to deal around with and then you download the model that you're looking to do. So here's a couple of things. So first they have them separated out into speech to text, text to speech, which is the voices. And they also have other, which is punctuation, which I'm not sure what that actually does. It may do the thing that I just said I wanted it to do with the punctuation. We'll download that and find out here in a second. So you find the thing that you want to download, you hit download. Now one note of caution here is that the larger the model base, the longer it's going to take to transcribe your note. So if you were to use the large English whisper model large. So if you were to use the English whisper to large here, this one here, it would take you quite a lot of time to transcribe even the smallest of notes because it's basically pulling from a larger pool of data. It would be more accurate because it has that has more data behind it, but it's going to be much slower. And also the larger the data pool, the more CPU intensive the program is. So you won't be able to hear this on the microphone, but even those one sentence notes that I just took spun the fans up in my computer. So I have a fairly powerful computer and even my computer makes some noise when it's doing this. So just keep that in mind. It is processor intensive. There is an option to turn on GPU acceleration. So I will probably try that later on. I haven't tried it yet, but theoretically that should take some of the juice off from the CPU and make it a little bit faster. So what you'll need to do if you want to use this is download a speech to text. That's going to take your spoken word and put it into text, obviously. And if you want to have the voice there for you to be able to have them read it to you, you can choose from different voices as well. So it'll show you if it's male or female, it'll show you the accent. So there's British. It also has a label for some of the names of the people who actually did this. They named the model. So I've been using Piper, this one here, and then they have Danny and Arctic medium, things like that. So they have different names for the model. So you download both of those and then you can both listen and read. So that's the basic note taking functionality of it. There are some settings here that we should take a look at. So you can choose when it's listening. So there's another thing that I should actually say that I should have put there outlined that none of this is off your computer. Any of the data that you put into this stays right here. All the processing stays right on your computer. You download the models to your computer. You never have to transmit anything other than downloading the models and the voices. This application never sends any of your data to the internet. So if you're worried about privacy, you can you can obviously verify this for yourself if you want to if you have the tools to do so. But it doesn't take any of your data and send it anywhere else for more machine learning to happen. So just you can you can keep that piece of mind. So let's go back to the settings here. You can choose the listening mode. So basically this what this will do is when you press the lesson button, it will listen basically for one sentence and then it will process. So you don't have to press stop or anything. If you want to, you can change press and hold. So when you press the listen button, you'll talk and it will listen for as long as you're pressing the button. And then when you let go of the button, it will then go on to process. Or you can turn it to be always on. I don't recommend that. But if you hadn't had a need for it, it is available. The text depending style, you can choose to have it either immediately appear right after the last word of the same line or appear on a line break. You can hit restore punctuation, which will require that punctuation model that we just downloaded. And I believe that this takes, yeah, this does take up much more memory and CPU intensive workflow. So I just know that that would be slower if you had that enabled the GPU acceleration here, I'm assuming would make things much faster. I haven't tried it yet. And then you can also speed up or slow down the text to speech. So that voice that reads things to you, you can make it slower or faster. You can change the interface so that it looks a little bit different. And you can change the location of the language files if you want. Those are all the settings that are here and it's fairly simple, but it gives you some control over how this thing works. Now, the other thing that this does is translate. So it'll take your notes that you have in the other pane, or you can type in new things here into this box, whatever you want to do. And you can then translate them into another language. So obviously you would have to download the language that you want to translate to. So it would translate, in this case, to Spanish. And then you could actually have it read it. Now, one thing I noticed, and I'll play this here in a second, is that the Spanish voice that I downloaded wasn't nearly as good. Now, they didn't have as many options for Spanish voices. I'm assuming that's something that will fall out over time. It wasn't horrible, but it wasn't as good as the English one, at least in terms of quality. It sounded a little robotic. So let me play this for you. That time it actually failed. So I'm not sure why it just said error. So I'm going to go ahead and delete this first line here. And we'll read this one here. Error, speech attack, and initialization failed. And never did that for me before. It worked earlier. So let me actually go ahead and just do this. We'll translate that. That should be really easy for them to read. Yeah. It just, let's type something a little bit. We'll try this. So it's a little bit longer. Hola, cuál es la mejor pizza? So if you listen closely there, I'll play it again. Hola, cuál es la mejor pizza? It sounds a little robotic. I don't know if that's just the nature of this language pack or if it's this particular voice. I'm not sure. Maybe some of the other voices in the Spanish pack are better. I didn't download, I just downloaded this one. So that's one thing that I noticed. Another thing is if you go to, let's just go to Spanish, you'll see that some of the text to speech, like there's not nearly as much text to speech here as there is for English. I think that's pretty much obvious. So I don't know how many voices have a lot of options for text to speech. But they are available in many different languages like you see here. So you can get, you know, Esperanto and French and Italian and Russian and all this stuff. So you can see that there's a large number of languages here. So if we just click on another one, so we would click on Italian, you can see that there are a few text to speech for Italian as well. So it does the translation. You can either translate the notes or you can translate things that you want, that you've typed right in this thing. You can also, if you want to, you can save the audio file. So you'll convert the text to audio and then save as a wave. You can also, if we're in this other notepad, you can actually transcribe and audio file. So if you have a recording of yourself just talking from your phone or whatever using the record application, you could take notes with that. You could transfer that file to your computer, however you wanted to do that, and then have speech note transcribe that audio file into notes. Now, here's one thing that it's missing. And this is by far my biggest complaint. And it's really weird. I'm not sure why this feature doesn't exist because it's a note-taking application, right? So you see I have a note here. How do I save this note? So I've taken a note. I obviously want to have that note around for the next time I need the note. And how would I possibly do that without a save function? Now there is a saved audio file, so I could convert the text to audio and then re-transcribe it later, I suppose. But if I want an actual text note, how do I save that? Well, I would copy it, go to my favorite text editing application, paste it, and then I have my note saved. That is a huge, huge miss on the applications part. Just allow me to save this as a text file and we'd be good, right? That's all I want to do is save this as a text file because if I'm talking to this thing and I'm taking notes, I want to have a way to get that text out of the application besides copying it and manually pasting it into another application. That's just silly. I don't understand why that feature's not here and it's just kind of mind-blowing that it's not. Now it's possible, as is usual with these things, that I've just missed it. But I don't know where I could have possibly missed it too. There's only two items in the file menu and that's it. None of these buttons here do anything. This is clear, this is undo, this is paste, this is copy, and that's all there is to it, right? There's not anything else. So I don't know where I'd possibly miss it too. It's not in the settings. As far as I can see, you change the interface style, you change the font size, but it doesn't have anything here about saving your notes to a text file, which is I think the primary goal of a note taking application is to save the notes somewhere, but this can't do it as anything other than an audio file. That's really weird to me. The technology here is super cool. I like that it's using whisper. I think whisper is very powerful. And if you want to have speech to text in a note taking application and you just wanna sit here and you wanna talk and you wanna have it transcribe automatically, this is one of those things that does a really good job for that. I wish kind of that it did it in real time so I could be sitting here talking and it would just kind of put this, the text right here in front of me as I was talking, but it doesn't do that. It requires you to talk, make it stop listening in some form or fashion, either having it do it automatically or whatever, and then it will process, right? Now obviously doing things like that in real time would require a ton of processing power so I can see why they don't do it, but that would be cool as well, but it doesn't have that feature. Honestly, the whole save to text thing is the biggest missing feature here. So if you are going to use this, just know going in that you're going to have to take the notes out manually and put them somewhere else if you wanna do that. If you're just using this as a transcriber feature or a transcriber application where you're transcribing your real time voice into text, you could use like that and then maybe you don't need that text out of there or you're perfectly willing of copying it out and putting it in wherever you need to go. That's fine. I would just like that to be an option. Just let me save it as a text file, that TXT. It doesn't need to be anything special. It can possibly be that hard. They've done the hard part. They just need to take it one step further and it would be awesome. So that is speech note. It's a really cool application that has a few flaws, but I think it's fairly new. I don't know exactly when it was released or anything like that, from what I've read, it sounds like it's fairly new. So maybe the features that I'm looking for will come eventually. But overall, I really, really like it. And if you, like I said, if you need to have text to speech, speech note is a really good option for that. So if you have thoughts on this, leave those in the comment section below. I'd love to hear from you. You can follow me on Massage Honor Odyssey. Those links will be in the video description. If you haven't already, leave a thumbs up on this video. It really does help the channel. I mean, seriously, it helps the channel. So thank you so very much for doing that. You can support me on Patreon at patreon.com slash linuscast. I'm also on kofiakio-fi.com slash linuscast. There's also a merch shop if you wanna go check out some of my merch. That's at shop.thelinuscast.org. We have t-shirts and mugs and a whole bunch of other stuff. So go on and check that out if you wanna support the channel. Thanks to everybody who does support me on Patreon and YouTube because we're all absolutely amazing without you, the challenges would not be anywhere near where it is right now. So thank you so very, very much for your support. You guys are awesome. Thank you so much for your support. I think I said that a few times but it bears reiterating many times because I'm very, very thankful. Thanks everybody for watching. I'll see ya next time. I'm really bad at endings, but everybody knows that. So also, what is the best pizza? That's a trick question. All pizza is the best.