recoder ces tiktok débiles pour voir si ça paie

Summary

Highlights

• Using Python and Pygame to replicate popular TikTok ball videos. 🏀
• Humorous take on simulating physics and gravity through coding. 🍃
• Challenges of integrating melodies in videos with coding and existing AI tools. 🎵
• The struggle for TikTok monetization and viral content creation. 💸
• Learning the importance of catchy titles and subscriber engagement on social media. 📣

Key Takeaways

• Coding can turn mundane concepts into potential viral TikTok content! 🤖
• TikTok's algorithm may favor quantity over quality in content creation. 📈
• Combining humor with technical skills can make for engaging tutorial content. 😂
• Perseverance and adaptation are key in the ever-evolving world of social media. 🔄
• Turning complex coding tasks into simple, step-by-step processes can make technical concepts more approachable. 🛠️

Overview

Chapters

• 00:00 - 00:30: Introduction The introduction begins with a casual and somewhat humorous tone, questioning who might be watching the video and expressing surprise. The speaker playfully engages with the audience, suggesting that if they aren't subscribed to their secondary channel, they might be 'horrible.' The segment ends with a joking demand for the audience's cell phones, maintaining a lighthearted and engaging interaction.
• 00:30 - 01:00: Exploring TikTok In the chapter 'Exploring TikTok', the focus is on the addictive nature of TikTok, a popular social media platform. The narrator comments on how TikTok videos, often considered empty in content, rapidly gain millions of views and offer minimal financial reward, estimated at around 400 euros per million views. The idea of creating an automated program to mass-produce such viral videos is also entertained, highlighting the ease with which content can attract massive audiences without substantial substance.
• 01:00 - 02:00: Coding the TikTok Video Program In this chapter, the focus is on creating a TikTok video program using Python, moving away from using languages like C and assembler. The narrative captures the author's enthusiasm for Python's simplicity and usability. Despite its lack of optimization when compared to other languages, its efficiency in ease-of-use is highlighted. The code will be built around the concept of creating a basic video game using Pygame, a Python library commonly used for game development. This game will involve simulating 'ball videos', and the process will begin with setting up a screen.
• 02:00 - 03:00: Implementing Gravity The chapter titled "Implementing Gravity" discusses the initial steps of a programming tutorial, where the author begins by setting up a digital canvas sized 1080 x 1920 pixels and fills it with black. They proceed to create a red ball, represented as a circle with a 40-pixel radius, placed in the center of the screen, complete with a meaningful outline. The author humorously reflects on the simplicity of the task compared to previous endeavors, particularly highlighting the ease of using Python over past experiences like explaining a BIOS-level 'Hello World'. The narrative briefly mentions the foundational physics of gravity, referencing Newton, as a playful nod to the concept of an object falling when dropped.
• 03:00 - 04:00: Simulating Realistic Physics The chapter discusses the simulation of gravity in a digital environment, focusing on handling the position and speed of a ball. It emphasizes the creation of a class to manage these properties effectively. The aim is to produce a video lasting 1 minute and 1 second at 60 frames per second, resulting in a total of 3660 frames, to meet TikTok monetization standards. Each frame involves rendering the ball in its calculated position.
• 04:00 - 05:00: Adding Bouncing Mechanics In the chapter titled "Adding Bouncing Mechanics," the concept of integrating gravity into a zero-gravity framework is discussed. Gravity is simulated by continuously subtracting a fixed value from the ball's vertical speed, effectively increasing the speed downward, thereby accelerating its fall. Initially at rest (speed 0), the ball's speed increases incrementally (-0.3, -0.6, etc.), and its position decreases correspondingly, demonstrating the effect of gravity. The explanation includes a brief video demonstration lasting 2 seconds, highlighting the ball's fall into infinity without external constraints for 59 seconds afterward.
• 05:00 - 06:00: Music and Melody Extraction The chapter titled 'Music and Melody Extraction' starts by diverting into an analogy involving gravity and physics, discussing concepts like terminal velocity and friction in the context of simplifying complex processes. The narrative humorously transitions to emphasize that while complex solutions, like killing a fly with a bazooka, may be effective, they are not always the most practical. The chapter highlights the importance of simplicity in processes, particularly when creating something as accessible as a TikTok video, rather than over-complicating the task at hand.
• 06:00 - 07:00: Educational Segment Ad The chapter discusses a scenario where a ball needs to bounce on an additional circle. It explains how to calculate whether the ball should bounce by measuring the distance from the center of the ball to the center of the circle. If the distance plus the radius of the ball is greater than or equal to the radius of the circle, a rebound should occur. The process involves some mathematical calculations to determine the new velocity (v') of the ball after rebound from its initial velocity (v). An unexpected interruption occurs briefly, but the focus remains on calculating the velocity after the ball rebounds from the circle.
• 07:00 - 08:00: Sound Frequency Analysis This chapter explores the concept of decomposing vectors in the context of sound frequency analysis. The tangent and normal lines at a point of impact on a circle are introduced as tools for understanding vector components. It is explained that the speed vector before and after rebound can be divided into tangential and normal components, respectively. Post-bounce, the tangential component remains unchanged, while the normal component reverses direction.
• 08:00 - 09:00: Melody Isolation Challenges The chapter explores the mathematical breakdown of motion, focusing on isolating melody challenges. It details the process of determining the tangential and normal components of a vector, particularly in the context of a moving ball. The computation involves subtracting the normal component from the vector, a critical step to isolate the tangential component. Furthermore, it discusses finding the direction of the normal, achieved by considering the ball's position at impact relative to the circle's center and normalizing it to get the precise direction. The explanation concludes with a practical application of these vectors in understanding motion.
• 09:00 - 10:00: The Final Approach to Music The chapter discusses the mathematical computation for the speed of an object (such as a ball) after a rebound. By projecting velocity onto a normal and applying a formula, the speed post-rebound is achieved. The main focus, however, is on using rebounds for an audio-visual experience where each bounce plays a note of a recognizable melody, enhancing viewer engagement. Future improvements are hinted at, but the current highlight is the blend of physics and music using bounces.
• 10:00 - 11:00: Enhancing Visual Content This chapter discusses the challenge of extracting notes from the main melody of a wav song, highlighting the technical difficulty involved. The narrator humorously admits their lack of knowledge in this area and suggests a need for further education. They mention DSTI as a school option in Paris or the Côte d'Azur, featuring arcade games and a dog mascot, before humorously suggesting that this promotional segment could end without further elaboration.
• 11:00 - 12:00: Gamifying the Video The chapter explores the features of the DSTI Bachelor of Science program, emphasizing its focus on English language instruction, which enhances vocabulary significantly—more than Shakespeare's, humorously suggested. The program offers flexibility for internships or work-study options either in France or internationally, making it attractive on a CV for the global job market. With a connected network of 400 employer companies, employment opportunities are ample. Graduates earn an RNCP level 6 title recognized by the State, along with industrial certifications and the option to pursue further studies in Data Engineering or Cybersecurity.
• 12:00 - 13:00: TikTok Posting and Experimentation This chapter discusses a range of advanced technology topics including computer architecture, networks, cybersecurity, AI, Cloud Computing, and DevOps. It highlights the availability of personalized support and modern tools for learning, as well as additional training courses ranging from Bac plus 3 to Bac plus 5. The chapter also introduces the YIN algorithm through a musical analogy, explaining how a guitar string vibrates at a certain frequency to produce a note.
• 13:00 - 14:00: The Viral Strategy This chapter explains the concept of sound waves and how they relate to music. It describes the movement of molecules in the air when a sound is produced at a frequency of 440 Hertz, commonly known as the musical note A. This movement creates a pattern of high and low-pressure areas. The chapter illustrates how this pressure variation over time can be visualized as a curve, which is essentially how music is represented scientifically.
• 14:00 - 15:00: Monetization Insights This chapter explores the concept of sound and frequency, using the example of a 440 Hertz note. The method of finding repetition by shifting signals is explained. Despite the potential for chaos and lack of clear repetition in real scenarios, the chapter emphasizes the existence of a note and the challenge of identifying the hidden frequency. The process involves starting with an initial function and adjusting to uncover the frequency.

recoder ces tiktok débiles pour voir si ça paie Transcription

• 00:00 - 00:30 ... Who even watches this? What ? ... You're kidding me! ... ... ... Oh, it's been a while! What, what? You missed me? Awww, it's not mutual. Hey maybe you're horrible! And you probably are since by saying that, it tells me you're not subscribed to my secondary channel! Okay, stop complaining and give me your cell phone.
• 00:30 - 01:00 Today we are going on TikTok. You have no idea what's going on, huh? Me neither. This platform from hell produces videos that take us straight back to our first hours on Earth. Tickles our dopamine receptors just right to create an addiction to empty content. There are these videos with a ball bouncing in circles that get millions of views. And on TikTok, a million views is like 400 euros. Imagine creating a program that mass produces them!
• 01:00 - 01:30 I don't want to imagine anymore. We're going to code that. And in Python this time. We're done with C and assembler. It's time to chill, to rise to heaven. The smoothness of the best programming language of all... And awful optimization capabilities, eh! We're going to make ball videos. What are you talking to me about optimization? Shut up. And admire the simplicity, man. We will use Pygame, a games library. Yeah, basically, we're going to create a video game with our balls and record the screen to make our videos. First, we need a screen.
• 01:30 - 02:00 1080 x 1920, fill it with black. Boom. A ball now. Just a circle. In red, in the middle of the screen, with a radius of 40 pixels. Okay, perfect, that. A little outline would be nice, though. For this, we actually draw two balls. The first one is just the outline, then the second as before. And there you have it, it's already done. Holly puppy, it took me 8 minutes of video to explain how to do a Hello World via BIOS last time. Oh, Python, I missed you so much. But yeah, we're sleeping a bit here. We'll have to ask advice from brother Newton, famous for discovering that if you drop an object, it falls.
• 02:00 - 02:30 Damn, the standards of the time suck. Well, to simulate gravity, we need to note somewhere the position of the ball and its speed. So we're going to create a class for our ball and put all that in there. We also want our video to last 1 minute 1 second. For TikTok monetization, you know. We are at 60 fps, so we will have 60 x 61 seconds equal to 3660 images to generate. We will go through these images and draw for each one our ball
• 02:30 - 03:00 This is for the zero-gravity version. To add gravity to each frame, we subtract an arbitrary value from the vertical speed of our ball. After doing this, we remove the speed from the position of the ball, which means that our ball will accelerate more and more. Concretely, at the beginning, it is at 0. Then, its speed increases to 0.3. Its position drops by 0.3. Next image, speed of minus 0.6. Its position drops by 0.6 to reach 0.9 and so on. And here's our gravity! And here's also a video that lasts 2 seconds, then 59 seconds of nothing. Our ball falls to infinity.
• 03:00 - 03:30 So... First of all, the formula for simulating gravity is M times G. And then your ball will accelerate to infinity. While terminal velocity exists. That's right! After 500 meters of free fall, friction compensates for the weight of the object. We achieve a balance. The speed doesn't move... Who is this guy!? We're not doing a NASA simulation, but a TikTok video. Often the best solution is not the best solution. Killing a fly with a bazooka is very effective, but it requires a lot more resources than a fly swatter. We're not going to complicate our lives for nothing!
• 03:30 - 04:00 Well, something should happen. We are therefore going to add another circle on which we would like to bounce. We determine if we should bounce by calculating the distance from the center of the ball to the center of our circle. If this distance to which we add the radius of the ball is greater than or equal to the radius of our circle, we must make a rebound. And how do we make a rebound? We're going to have to do some math there. Wtf is he doing here? Get out! OK, we have our ball which will bounce on our circle. We already know its velocity, which we note v, and we want to find the velocity after rebound, which we note v'.
• 04:00 - 04:30 To do this, we will draw the tangent of the circle at the point of impact, as well as the normal. The tangent is the line that touches our circle at a single point at the level of impact, and the normal is the line perpendicular to the tangent at the level of impact. The speed vector before rebound can be broken down into two vectors. Its tangential component, parallel to the tangent, plus its normal component, parallel to the normal. And after the bounce, the ball keeps the same tangential component, but its normal component is just reversed.
• 04:30 - 05:00 So it's equal to the tangential component of v minus the normal component of v. The tangential component can be replaced by v minus the normal component of v. We therefore obtain v' is equal to v minus normal component of v times 2. V, we know, and the normal component of v, we can know it. Okay, now we need to find the direction of the normal and that's simply the position of the ball at impact minus the center of the circle. We know all that. And we can normalize it to get the direction without it impacting anything else, and we call that n.
• 05:00 - 05:30 Then we project v onto this normal via this formula. We therefore obtain that the speed after rebound is equal to the speed before rebound minus 2 times the scalar product of v and n multiplied by n. We have all these values. We can make the rebound by replacing the velocity of the ball with this. Here it is! Later, we will improve a lot of things, but we have the most important. And before we finish all that, we're going to tackle the most important piece. Play music with bounces. All these videos capture the viewer's attention because with each bounce, it plays a note of a well-known melody.
• 05:30 - 06:00 Firstly, we need to be able to take a wav song and extract all the notes from the main melody. And that is tough! I don't know how to do it. I guess I need to go back to school. Refresh my memory on all this. But we don't go to just any school. We're going to DSTI. Campus in Paris or on the Côte d'Azur, sunshine 365 days a year. You can play games, at the arcade machines, they even have a dog mascot. Look at him! Really, I can stop the ad segment there, right? I mean, what else do you want me to tell you?
• 06:00 - 06:30 Oh ok. Monsieur is the serious type. Monsieur wants to study to learn things. But that’s good. I present to you the DSTI Bachelor of Science program. As the name suggests, everything is in English. You come out of there with more vocabulary than Shakespeare. And what's more, you can do an internship or work-study program in France or abroad. Perfect in a CV for the world of work, that. Either way, with a network of 400 employer companies, you will find work. You also come away with an RNCP level 6 title recognized by the State, industrial certifications, and the possibility of continuing with a master's degree in Data Engineering or Cybersecurity.
• 06:30 - 07:00 You will learn computer architecture, networks, cybersecurity, AI, Cloud Computing and DevOps, with personalized support and modern tools. They also have other training courses available from Bac plus 3 to Bac plus 5. I'll put them there for you. And if you want more information, don't hesitate to go to the link in the description. Ok, I know exactly what we need now. The YIN algorithm. And no, it's not a joke. The name literally comes from that. Let's imagine that I play a note. With a guitar. The note is played thanks to the string which will vibrate at a certain frequency.
• 07:00 - 07:30 440 times per second, for example, or 440 Hertz. As it moves, this rope will push molecules in the air which will imitate this movement. But also spread it by pushing the following ones. If we place ourselves at a specific point, for example here, we can observe that we have a series of moments where there are a lot of air molecules in the same place, then practically no more. So we have high pressure, then low. If we note the pressure at this location over time, it gives us this curve. And that’s how we represent music.
• 07:30 - 08:00 Or the sound. The note played will depend on the frequency with which we cycle. So, I said it earlier, here it's 440 Hertz. That matches the note there. But if we didn't know, how could we find her? Well, you just have to count then. We place ourselves there, and we shift our signal little by little until we find how many seconds we need to shift to have a repetition. Okay so here, this is the perfect case. In reality, it would be more like that. No clear repetition, chaos. But there is a note. You just have to find the frequency hidden in there. To get there, we take our initial function,
• 08:00 - 08:30 and we shift it by a certain time, tau. On an arbitrary window, we look at the difference between the initial function and its shifted version, and we sum them. We calculate this for all tau, and if we look at the results, we see that the minimum for this sum is reached at regular rate intervals. This is our hidden frequency. So according to the paper on the Yin algo, it would be better to take the cumulatively normalized version of this function. But I think it's useless. So take this with a grain of salt, bigger. Even bigger. Because the source is me.
• 08:30 - 09:00 But I tested lots of signals, and each time, with the non-normalized difference, that's enough. And the paper indicates that we would go from 1.95% error to 1.69. So yeah, it must be hard to find counterexamples. But anyway, after implementing all that, it works really well on melodies played by a single instrument. But the problem is that the music is played by lots of instruments at the same time, which play different notes. And our code, well, it can't know which note to focus on.
• 09:00 - 09:30 And the result is catastrophic. To successfully isolate the melody and focus on it, artificial intelligence must be used. And I found this tool from Google that allows you to do it, and the result is much better. But really... It's crap! Except I don't think I can code something better than Google devs. So, I gave up.
• 09:30 - 10:00 And in the end, I simply search the Internet for famous music in MIDI version. You know, MIDI is... That kind of sound. Wait, wait, wait, wait. Does that mean? All this research? For nothing. Weren't you the one who said not to complicate things? I don't want to talk about it. Okay, back to the code. I have .MIDI files. I can retrieve the individual notes, and each time we bounce, we play one. Now we have to find a way to make this... Interesting. All the content that went viral with these videos have one thing in common. Several circles. Not a single one. With rotating holes.
• 10:00 - 10:30 And the ball tries to escape, but never succeeds. Obviously. Okay. Let's code this. We move our circle. Now it will be an arc. Just to get the hole. It is also made to spin on itself at a certain speed. If the ball is in the hole, it does not bounce, but it breaks this circle with some visual effects to keep the audience's attention. Good. We need more circles. I'm creating 1000 of them. With increasingly larger radius to stack them. We only draw those with a certain radius to avoid wasting resources. We are slowly reducing their radius
• 10:30 - 11:00 so that we always have them on the screen as well. And we're not bad there. But there is no point in watching our content at the moment. We are going to add an issue. For this, we need two balls of different colors. As soon as a ball breaks a circle, it gains a point and the result is displayed on the screen. Why? Well, you know Paul the octopus? I told myself I was going to do the same. Predict match results. Who will qualify? PSG or Arsenal? And now you want to know. You stay. I believe it can do it. Everything is ready. All I have to do is create a TikTok account,
• 11:00 - 11:30 post and flop. I think it's because TikTok shadowbanned my account. Perhaps the description was in poor taste. Okay, let's delete and start again. Come on. A few hundred, sometimes a thousand views. Not crazy. So I started experimenting. Use memes maybe Flop. Play on a controversy. Flop. But then I posted this video. "Are you stupid? Respectfully."
• 11:30 - 12:00 Oh yeah, we're not influencers for nothing here. Two observations. Number 1. Almost no subscriptions per 100,000 views. Need to be fixed Number 2. People share content. They send it to their friends like, “Yeah, look, you’re stupid!" That's why it went viral We are going to combine these two points by taking people's comments to make the title of the video and clearly indicating that you also have to subscribe to be selected and have a response. It's fake btw I don't check. And in doing that, it went viral. In one week, 2 million views, 850 subscribers.
• 12:00 - 12:30 And how much would that bring in? Well, you need 10,000 subscribers for monetization. But based on the numbers from my main account and knowing that around 63% of views are long enough to be eligible, we'd make almost 800 bucks in a week. If you are self-employed, that would make you almost 2300 euros net per month. I got disgusted. My main account, never in life, he does this on TikTok. And I find that a shame. It's not so much the quality that counts It's often the quantity. You have to post a lot If you never hear from me again one day, it's probably because I decided to make videos for 8-15 year olds.
• 12:30 - 13:00 Well not me, my robot.