How to Use Bark Ai: FREE Text-To-Speech Tool

in today's AI tutorial I'll be showing you guys bark the brand new text-to-speech pattern recognition model which can not only perfect text-to-speech but also the variations in your voice before we head over to the PC setup to show you how to install it and how to use it if you guys could go down and leave a like on this video And subscribe to the channel if you like AI related content I post daily videos on different ways you can utilize AI in your daily lives with that being said let's head over to the computer right now alrighty so now that we're on the PC setup I'm going to be showcasing all the cool stuff that comes with the program bark and why it's actually better and then 11 labs are going to do a side to side comparison of the two programs now there's actually a lot of reasons as to why bark is currently in my opinion the best text to speech AI model out there right now and that's just because it can make your voice a lot more variable so what I mean by this is that in most a AI voice models that we've seen before it talks very monotone and doesn't pick up when text-to-speech has like brackets around words that give tone to words so for example if I wanted to say I'm very mad right I said that in a very angry tone however when you do Texas speech in some of these older models it doesn't pick up on that anger and we'll just say I'm very mad so this is what makes Sparks Stand Out is that you can really pone into your voice and it makes the output so much more realistic which is so cool additionally bark supports multiple languages which just wasn't a thing with models in the past which is really cool so here's an example showcasing the model with a Spanish language buenos dias Miguel Milo but I suppose your English isn't terribly so as you can see I don't know a lick of Spanish but it sounds pretty decent and here is an example of the model being used with the different variations so as you can see it says hello my name is suno and uh and then obviously it's a pause because it would be a stutter which is very normal in the regular person's speech patterns and I like pizza and then in Brackets laughs but I also have interests such as playing Tic-Tac-Toe so let's see what it's able to do with these tones indicated hello my name is suno and uh and I like pizza but um I also have other in so as you guys could hear from that text to speech that sounds almost perfect there is a little bit of that robotic static but otherwise it sounds amazing and you can even do it in like a sing-songy way like like in this prompt right here that says in the jungle the mighty jungle the Lion's bark tonight but it has these music notes around it so it's going to say it in that sing-songy way that's like in the jungle the mighty jungle so let's hear that play out [Music] tonight so as you guys can hear it says it in that sing-songy voice obviously not that very well I will say that out of all these features the music portion is by far the worst but it still is cool that it can pick up on the fact that it wants you to do Texas speech in a sing-songy way now not only can bark do Texas speech it can also do audio and voice cloning and it's actually better than 11 Labs which so far has been the leader in the space in my opinion of AI voice cloning so that is very cool in addition to this you can actually set up different speakers so let's say you wanted to do a whole podcast script where it goes back and forth between a guy a girl a guy a girl and they keep talking back and forth well you can set up romps so that there is a narrator there's a man there's a woman and you can see this in this example right here please wow that's expensive so pretty cool stuff it's very distinguished who the girl is who the guy is and what they're stating I will say there is still some of that like robotic-ish voice you kind of hear that's a little eerie and doesn't sound totally human but remember this is the worst it's ever going to be it is only going to get much better so let's do that side by side comparison of the two models we're actually going to be using the hugging face space right here I'll have the link down below if you want to generate your own audio without actually downloading the program at all to your computer but do be warned that there are a lot of people trying to use this right now and it is going to take you a long time to generate audio like over a hundred seconds so that's like what a minute and a half or a little over a minute and a half just to generate One Singular audio so if you just want to test it out once or twice definitely use the hugging face but I do recommend installing locally to your computer or using the Google collab which I'll show you right after we do this comparison all right so I have my prom right here that says hello my name is AI Kingdom and uh I want you to subscribe to the AI Kingdom but I also want you to leave a like thank you and then we'll add something in to make it fun we'll say cry so first they laugh then they cry so I'm going to submit that and as you can see it's gonna take around 106 seconds here to create our audio so we're just gonna have to wait for that to be done but in the meantime I'll show you guys my 11 labs and we have almost The Identical prompt however in 11 Labs you can't even do things like add in those brackets that says laughs or cries so we just have to keep it based like this so it's going to State the exact same thing and I'm going to hit generate and we're going to see what it sounds like we just are doing it on one of the pre-made people we'll just do Adam here and I'm going to select generate so as you guys can see it is not horrible but also some things one it didn't pick up on AI it literally said welcome to the eye Kingdom like it just ignored a for whatever reason I guess if you wanted to do a workaround to this you could do a and then I like this and it would probably pick it up but this is just one of the examples showcasing why bark is exploding right now because of how much better it is than literally every single other voice to speech program and we can even try on my own voice that I cloned we'll see what this sounds like hello from the AI Kingdom and uh and I want you to subscribe to the eye Kingdom but I also want you to leave a like thank you so I don't know how much that sounds like me I wouldn't really say it does but as you can see it's still incredible but bark is doing it so much better and this is the worst it's going to be so I will have the GitHub link down below if you did want to download it locally to your computer or there will be the collab which is much easier which we're going to use but before we get into that our audio is almost finished processing it's actually going over the 115 seconds so it actually took 130 seconds so even longer to generate and we finally have it so let's hear this audio hello my name is pie Kingdom and uh and I want you to subscribe to the eye King [Laughter] thank you already so as you guys can see the audio it generates it has lots of variation which is great which is what we're going for but absolutely creepy I don't know what it is with bark right now but I find all the audios they generate are just Eerie like they sound spooky maybe it's like the creepy laugh I added in also I don't think he cried at the end of the audio so we'll have to see why that was but I'm gonna figure that out off the tutorial let's go into collab and show you guys how to install everything it is so easy to do all you want to do is make sure you're logged into your Google account at the top there and then you just want to keep clicking these run cell buttons or these play buttons If you haven't used Google Cloud before it's very easy that's literally all you have to do so as you can see we installed all of this now the next thing to do is go to the basics and you just want to click the play button again and it is going to install everything now while this is installing I'll show you how you can actually edit the text prompt and here is the prompt right here so we're it says text prompt and all you have to do is type whatever you want to type in so I'm just going to copy our prompt from earlier and I'm going to paste this in here alrighty I am back I did have an error and all I had to do was just re-run the cell and then it worked so if you do come into that issue all you have to do is just reinstall and it should work for you but here we got our prompt that is hello my name is AI Kingdom blah blah blah blah blah so we're going to select the play button here and it is going to generate our audio as you can see going a lot faster than the hugging face space so that is the benefit to using Google collab instead and also it doesn't have to install anything on your PC which is probably the nicest thing but if you did want to choose to do that you can head over to the GitHub repo and I'll have that linked down below and there we go our audio was created it only took 54 seconds which it was a lot shorter than the hugging face option so we're going to play this right now hello my name is I Kingdom and uh and I want you to subscribe to the eye Kingdom but I also want you to leave a like thank you alrighty so that is our audio generation from bark and a pretty decent job although I don't know what it is but with this audio it also didn't understand AI however in the hugging face speaker they knew that it was AI so that's very interesting I don't know why they just assume it's like I that's how they pronounce it but the laugh was on point even the thank you extending that was on point as well and being very enthusiastic so overall I'm pretty happy with this and if you do want to download it all you have to do is right click and you can download your audio so this is going to change the game obviously once these cloned voices sound a lot better and it is going to be up to you to figure out which model is going to be best for the specific audio you're going for but yeah that's just a quick little rundown on how to use spark how to install it in the cool benefits with it if you enjoyed this video make sure to leave a like on it subscribe to the channel Channel and I hope you guys have a great rest of your day

Channel: Trent Kingdom

Views: 23,362

Keywords: tutorial, ai, ai tutorial, ai news, Bark AI, Bark text to speech, voice cloning, TTS tool, AI tutorial, free TTS, Bark tutorial, Bark voice cloning, clone voices, voice synthesis, AI voice generator, Bark AI guide, text to speech AI, free voice cloning, Bark AI review, Bark voice synth, custom TTS voices, AI TTS, Bark tool tutorial, create AI voices, voice cloning software, Bark AI tips, Bark AI tricks, voice generator tutorial, how to install bark, how to use bark

Id: p1dlZZo8WjU

Length: 10min 37sec (637 seconds)

Published: Wed Apr 26 2023