Transcription
[Music] Hello everyone. This is Professor Patterns and in this video we're going to be covering the 11 Labs V3 Alpha model. Now, a lot of people have been talking about this one. Uh, I have some people commenting in some of my earlier videos. Um, I will say I've never been a huge fan of 11 Labs before, just because it's a paid service, especially for text-to-speech, and there are so many like free services such as Kokoro and Orpheus text-to-speech available. Uh, but I thought, might as well at least try it out. Um, see what all of the hype is all about.
So, the 11 Labs V3 alpha model, um, they did also share a sheet of different voice-related, um, things like panicked, tired, shouting, and stuff. There are some sound effects like gunshots, rainfall, and then some unique ones like strong accent. So, if I want like maybe a strong Italian accent or something, I guess I can put that. Um, I basically just put that into ChatGPT. I said, "Give me four sentences using some voice-related sound effects." And, uh, it gave me some sentences here. So, let me just copy the first one. And I'm going to paste that right here.
There are a couple of different voices that you can choose, but they had some that are like, they call them the best voices for V3. Um, so, we'll actually try some of these out. We've got some James. "I must not fear. Fear is the mind killer." Hey everybody, this is Juniper. "Hey, are you looking for a fresh and engaging voice for your podcast or social media? Then I'm the voice for you." Okay, that's the one I selected. And, uh, it's hopeful, it's upbeat, it's clear, uh, it makes sense. Let's try that voice. Turn up the volume a little bit. And, uh, what's a sentence? Excited. "You won the competition." Giggles. "That's amazing. Seriously, I didn't think you'd actually pull it off." Applause. Um, I started off with 103, 306 credits and it says it cost 25 credits to make. Okay. So, that's not bad. I will say, let's generate this speech and see how good it is. "You won the competition. That's amazing. Seriously, I didn't think you'd actually pull it off." Okay, that it was, it sounded good. Not better than like Orpheus, maybe. Uh, let's try the other one that generated. "You won the competition. That's amazing. Seriously, I didn't think you'd actually pull it off." That was the saddest applause. Um, but this one, I think this second one was a little bit better for sure. Um, it had the giggle. It had the half kind of laugh into the sentence like, "This is what I'm talking about. Seriously, it's amazing. Serious." "That's amazing." Yeah, that I don't know if that was intentional or not, but that was pretty good. Uh, let's try the next sentence.
So, this one says, "Curious, what are you saying? This treasure under the old lighthouse." Dramatic pause, thunder. Okay, so maybe I want something a little bit more dramatic for this voice. Um, let's try Liam. "Life isn't about finding yourself. Life is about creating yourself." A single rose can be. "Friends show their love in times of trouble. Life without love is like a tree without blossoms or fruit." Yeah, I think Harry makes sense. So, I'm going to pick Harry. And this one costs 28 credits. Okay, just generate. "Curious. Wait, are you saying the treasure is hidden under the old lighthouse? That's actually kind of epic." All right, a couple of things that went wrong. First, it read out "curious" for some reason. Um, I feel like the voice kind of changed. It read this. It did the pause, the thunder, which is okay. And then this was in a completely different voice. Or am I just tripping? Let me try the other other one that I generated. "Cursor. Cursor. Wait, are you saying the treasure is hidden under the old lighthouse? That's actually kind of epic." I still feel like it's changing its voice halfway through the thunder. Or am I am I wrong? Let me know in the comments if you think so as well.
Let's try the third sentence. Okay. Starts laughing. "Oh no, not again." Snorts. "You and raccoons have the weirdest beef." There's a fart. Uh, what voice did I choose for this one? Uh, maybe by repeating what students say. "Hey, you're not asleep yet, are you? Oh, come on. You think I don't see what's happening here? Please. I was two steps ahead before you even laced up." All right, I'm going to pick Priyanka Sogum. Late night radio, neutral accent. Um, let's generate speech. "How much is this? Is it matter?" Uh, okay. 19 credits. So, the farts are cheaper. So, that's great. "Oh, no. Not again. You and raccoons have the weirdest beef." Okay. Uh, the fart was a little bit underwhelming. Maybe let's go back to Blondie. I think Blondie was good. And let's generate again. "Oh no, not again. You and raccoons have the weirdest beef." I didn't expect the explo the explosive part um on there. "Oh no, not again."
[Music] "10 on 10. 10 on 10. I think it's following us. Don't make a sound." Wow, that was good. That was really good. The heartbeat, the dramatic effect, the underwater, the "don't make a sound." I mean, if I'm writing like a horror kind of audiobook, I feel like this is a great voice. But the good thing is that you can actually download, um, the entire voice file. And what is it? A, it's an MP3 file. Okay, nice. So, you can actually download the entire MP3 file. Great.
Let's try a full conversation. Maybe like multiple speakers and let's see how that goes. "Oh no, please, someone save me. I am in trouble. Um, help someone, please." And then here, let's add a villain. So, maybe for this villain voice, I'm going to pick something like Reginald. Intense villain. Um, "No one can save you here." Let's, let's add a fart in there. Um, and then maybe an evil laugh. What other tags are there? Explosion. Yeah, let's do that. Um, "Haha, explosion." And let me add another tag. Maybe something like applause. Um, "You are in severe danger." Applause. Let's add another. And this can be a follow-up by Blondie. "Please, someone save me again." And now we can have a hero come in. Maybe the hero could be Kuan. How about this one? "Now, if you're ever down this way, don't be shy. Yeah, this is the one. We southern folk love having folks over. Um, stop right there." And then maybe a gunshot. Um, "You, sir, are under arrest. Um, I'm going to take you back to the sheriff or I guess I'm the sheriff. I'm going I'm going to b take you back to downtown to the town to to the place in where they go, um, to prison." Um, and then add a speaker. And in this case, Reginald responds, "No, you will not." And then let's make him fart again. Um, "All right." Let's see how much this costs to make. 60 credits. Okay, so it does start racking up, but maybe it's not still not like a huge amount or anything. Uh, I think the subscription that I had was the creator subscription. That's, I don't have the pro one. I have the creator one. And this gives me 100,000 credits per month. If I want more credits, is that okay? So, that's 30 cents for 1,000 credits. So, that's the overall cost.
Let's go back here and let's generate the speech. "Oh, no. Please, someone save me. I'm in trouble. Help. Someone, please. No one can save you here. You are in severe danger. Please, someone save me. Stop right there. You, sir, are under arrest. I'm going to take you back to prison. No, you will not." I don't know what it is about that fart noise, but that one was okay. "Oh no, please, someone save me. I am in trouble. Help someone. What's with this song in the background? What? Oh no. Please, someone save me. I am in trouble. Help someone. Please. No one can save you here."
[Music] "You are in severe danger. Please, someone save me. Stop right there. You, sir, are under arrest. I'm going to take you." Why did that gunshot sound like a fart noise? "Um, back to prison. No, you will not." Um, okay. I, I like some aspects. I like the fact that the applause that's here, it carries onto the second part of the conversation. So, it's not like applause and then end and then it goes into the next part. So, I like some of those aspects. Um, what, what does enhance do? Adds audio tags to help guide the delivery. What do you mean? Oh, it adds the tags so you don't have to do it. Okay. Wait, it removed all of my farts. "Oh, no. Please, someone save me. Fart. I'm in trouble. Please, someone help." Um, let's, let's see what other tags there are. Um, there is a woo. Let's go with the woo. "I am in trouble. Someone save me, please. Woo." And then there is echoes. So, I'm going to put that in here. "No one can save you here. You." Maybe this can be an echo for sure. And then what about ASMR mode? I think that would be cool. Um, "Please, someone ASMR mode save me." I don't know how that's going to go, but it'll be interesting. And then lastly, maybe like a gulp that can come in after. "No gulp. You will not." How much does this cotton cost me? 77 credits. Okay, so I'm starting to rack up now.
"Oh no. Please, someone save me. I am in trouble. Help. Someone, please. No one can save you here. You are in severe danger. Please, someone save me. Stop right there. You, sir, are under arrest. I'm going to take you back to prison. No, you will not." That one was amazing. Minus the random gulp in there. But besides that, wow, that was that was good. Um, for an audiobook kind of material or something. I am a little bit impressed. I don't know if there is an open-source solution that comes close. The cost is weirdly or surprisingly not that bad. There has to be a catch though, right? Like, oh, it's got an 80% discount. No wonder. Okay, because this, this was good for how much I'm paying. Um, this was really good. But if it's at an 80% discount, okay, that's going to get a lot more expensive. Um, June 2025. What do, that's a month. Okay, so after a month, this model gets extremely expensive. Um, but you have a month to at least try it out on this discount. And honestly, not that bad. I mean, if you want to maybe create an audiobook in the month, um, go for it. Uh, but that's pretty much it for this video. Overall, not terrible. Still a paid solution, um, but not bad, honestly.
All right, that's it for this video. Thank you all for watching. I'll see you in the next one. Goodbye.