 oh no, no, no. Stun it over, stream. Just go over natethefilecast.com. That's what we do. If you have any feedback, if you check the site, and you wanna tell me what you think, this is a good time to do so. And why I can't note here myself. That's a bit odd, keep talking to test my mic. Okay it's working, okay now. to be using the correct microphone as well. That's good sign. So it's bionic chaos. For you, if you check the site, please do let me know some questions in various chats about what is it all about. Okay, the most recent tool was another tool, a blog, was this blind alt one plot. So by the way, full disclaimer, this was created by chgpt, generated some of the content here. Well, through many prompts, yeah, we generated this content, but I do stand behind every sentence. Well, kind of, there's still disclaimers just in case we made mistakes, but mistakes, but yes, it should be treated as text I wrote kind of thing. It would be on the YouTube channels, there will be videos of me generating this text. Probably should have. Yeah, let me know what you think. If I should add the video about it at the bottom of the page, for example, some people like we use a beta. What else we get? This motion particle, we haven't touched in a while. It's just too relaxed. But then, yes, there's actually a future direction of this one. How can I use the chgpt? I use it's a text to speech, has pretty good text to speech. That's the problem because I can't use it anywhere else, can I? Can you regenerate the same text? Not the reason I'm doing it. Okay, that was chgpt 3.5. I actually told it to repeat the same text, but it did not. Well, we have to regenerate. Can you just copy the text? It's not doing it, it's not just copying it, it's changing it. The reason I'm doing it is because I want that text to speech. Let's move on quickly. It's just a text to speech thing. This is one of the main applications that we have. We could just pop the link into gpt4. We'll assume it will just pick up on the text at the bottom of the page. Text to speech button go. That's about right. That's the main tool we were working on for a while. I don't think there's a need to go over this again unless you have any questions. There was a similar tool published a while back that one didn't have the robot option. It would just be playing against yourself. I don't know if we should get rid of it. Let me know what you think. I think it's fulfilling a different function there. Well, the new one is significantly different. First, you can add noise to the mix. It's happening between waveforms, not within the same waveform. So when you adjust this, it will only affect the next incoming wave like this. That looks abnormal. Yeah, the robot in this case should be starting making mistakes. Yeah, it got one false alarm there. When the noise is at the max, at the maximum. So let me know if you try this out. Do you think about it? Yeah, I can get gpt4, potentially giving a different perspective on this. Also, the fact that the machine is always a winning human. So that's not nice. Is that an image into gpt4 in the description? See what it says. Maybe it will come up with some new ideas to improve it or whatever. It's generating fairly slowly today. The board is a bit lazy. Sorry, I haven't done it on our screen because things just don't fit in. I guess I could replace it with this one. But we will start developing in a second. We have this other application, a neural network example thing that we will be trying to develop. Get to that in a second. Just read this one. Okay, that's pretty generic description. Shouldn't have clicked on that. Do I get a higher score than a human? Then a human. I am human, am I? I don't know anymore. I thought for a second there I had a higher score than the machine. But yes, if you rely on the machine, it had one miss. Do you know why? Check its algorithm. That's probably when I did the beta is when I didn't do the miss. The machine did. Now we have the same score. I think I missed something else, actually, not the same waveform. Oh, that's becoming a bit faster. I think there is a video of me reaching about like 1000 points and becomes crazy fast. So let me know if you tried this and what score you reached. There is currently no leaderboard or anything like that. Just self-reporting. This is also why doing the sessions is that I can get the, oops, missed that, can get the feedback on this stuff. And no, I don't have a C or anyone whispering in my ear. It's just me trying to check the sound. There's two pairs of headphones that are essentially the same shape. So it doesn't help. Because I always get confused which one is which one is which. Yeah, the sound should be okay. Anyway, the post is already like has doubled the points will because humans are really good at getting distracted. I wasn't paying attention to the waveforms. And if this was real data, the patient would have been misdiagnosed or like underdiagnosed, not diagnosed.