The ULTIMATE Guide to AI Voice Cloning: RVC WebUI (Zero to Hero)

[FurkanGozukara]

The ULTIMATE Guide to AI Voice Cloning: RVC WebUI (Zero to Hero)

Full tutorial: https://www.youtube.com/watch?v=ZRrzvD4wNys

Ultimate AI Voice Changer Tutorial: SECourses Premium RVC Web UI (Windows, RunPod & Massed Compute). This video is only for educational and responsible usage purposes. With V3: Multiple voice merge to generate custom voice feature implemented.

Welcome to the complete tutorial for the SECourses Premium RVC Web UI! In this video, I will show you how to easily transform your speaking voice or song vocals using our highly optimized AI voice conversion application.

Whether you want to sound like a famous celebrity (like Donald Trump or Tupac), replace vocals in AI-generated music, or change your voice live in real-time, this tool has everything you need. The installer automatically downloads 30+ pre-trained demo voices, and you can easily add hundreds more from Hugging Face!

I will walk you through the entire 1-click installation process for local Windows PCs, as well as step-by-step cloud deployment guides for RunPod, SimplePod, and our highly recommended cloud provider, Massed Compute.

✨ Key Features Covered in this Video:

• 1-Click Install with automatic model & dependency downloading

• 30+ Pre-trained demo voices included automatically

• Auto-separation of vocals and music for song processing

• Custom Preset System to save your favorite inference settings

• Batch Inference to process multiple audio files instantly

• Real-Time Live Voice Changing desktop application included

• Full Custom Voice Training support (VRAM usage below 4GB!)

• Fully optimized for Cloud GPUs (RunPod, SimplePod, Massed Compute)

🔗 Important Links & Resources:

📥 Download Installer & Full Instructions: [ https://www.patreon.com/posts/rvc-web-ui-app-installer-zip-file-149104996 ]

⚙️ Windows Requirements Tutorial: [ https://youtu.be/DrhUHnYfwC0 ]

☁️ Massed Compute Registration (Use coupon code SECourses for a discount!): [ https://vm.massedcompute.com/signup?linkId=lp_034338&sourceId=secourses&tenantId=massed-compute ]

☁️ RunPod Registration & Templates: [ https://get.runpod.io/955rkuppqv4h ]

☁️ SimplePod Registration & Templates: [ https://simplepod.ai/ref?user=secourses ]

⏱️ Video Chapters:

00:00:00 Introduction to SECourses Premium RVC Web UI

00:00:18 Pre-trained demo voices & Hugging Face repositories

00:00:54 Live demonstration of speech-to-speech voice changing

00:01:17 Processing songs and AI-generated music vocals

00:01:50 Overview of features: batch inference, real-time, and training

00:02:36 Downloading the installer zip file and reading requirements

00:03:06 Installing SECourses Premium RVC on Windows

00:03:41 Automatic model downloading process and verification

00:04:26 Where to place custom demo voices, weights, and indices

00:04:55 Starting the Web UI application and loading models

00:05:24 Auto separate vocals and music feature explained

00:05:38 Using the inference preset system to save settings

00:06:08 Uploading audio files and testing default voice conversion

00:06:37 Converting speech to Donald Trump demo voice

00:07:18 Adjusting parameters to improve voice conversion quality

00:07:37 Testing voice conversion on an AI-generated song

00:08:12 Converting the song vocals to Tupac demo voice

00:09:24 How automatic music and vocal separation works

00:09:56 Custom voice model training overview and FAQ

00:10:28 How to use the Batch Inference feature for multiple files

00:11:37 Accessing generated output files and their metadata

00:12:04 Real-time live voice changing desktop application setup

00:12:47 Live real-time voice conversion demonstration

00:13:13 Cloud deployment introduction: RunPod, SimplePod, Massed Compute

00:13:46 SimplePod and RunPod interface and pricing comparison

00:14:00 Setting up RunPod template and persistent volume

00:14:53 Important: Selecting the correct CUDA version on RunPod

00:15:47 Deploying the RunPod pod and waiting for initialization

00:16:13 Connecting to Jupyter Lab and uploading the installer zip

00:16:32 Running the installation commands on RunPod terminal

00:17:08 Starting the application on RunPod with Gradio Live link

00:17:33 Testing voice conversion and downloading files on RunPod

00:18:53 How to use RVC Premium on Massed Compute cloud provider

00:19:22 Deploying a machine on Massed Compute with SECourses image

00:20:41 Connecting to Massed Compute using the ThinLinc client

00:21:28 Transferring files using the shared thin drives folder

00:22:03 Extracting and installing the application on Massed Compute

00:23:15 Starting the application and launching Gradio Live on Massed Compute

00:23:54 Testing voice conversion and downloading outputs on Massed Compute

00:25:08 Deleting cloud instances to save credits and money

00:25:37 How to install and use the application on SimplePod

00:26:33 Speed comparison between SimplePod and RunPod installations

00:27:17 Conclusion and final thoughts

👇 Don't forget to LIKE, SUBSCRIBE, and turn on notifications for more amazing AI tutorials! If you have any questions, leave a comment below!

Video Transcription

• 00:00:00 Greetings everyone. Today I am going to  introduce you to SECourses Premium RVC  

• 00:00:05 Web UI. This application is used to change  voices, so you have an existing voice and  

• 00:00:12 you can change it into a trained voice. With this  application, I am downloading automatically 30+  

• 00:00:19 pre-trained demo voices. However, you can  find hundreds of previously trained demo  

• 00:00:26 voices in these 2 repositories of the Hugging  Face. You can use this search icon to search,  

• 00:00:32 like let's search for Trump, and you see there are  some Trump voices. You can also scan this repo and  

• 00:00:40 perhaps search from here. Let's see, no, let's say  Tupac. Yes, so you see you can also use the search  

• 00:00:48 bar here. And then you can change voices. What  I mean by that, for example, I have a demo voice  

• 00:00:55 here. Let me demonstrate you live. "Welcome to the  LTX 2.3 presets tutorial. In this tutorial..." And  

• 00:01:03 now let's see the changed voice. "Welcome to the  LTX 2.3 presets tutorial. In this tutorial..."  

• 00:01:12 This is not the best quality that you can get,  but this is a demo demonstration. Moreover,  

• 00:01:17 you can even process songs. I will show you  with a song that I have made recently with  

• 00:01:22 our newest song generation application,  which one is this one. Let me show you.

• 00:01:34 Okay, this was AI generated. Now let's see the  changed voice. So as you can see, it is working  

• 00:01:47 really good. We also have batch inference and a  lot of features. I will show every one of them,  

• 00:01:52 how to install and use this application on your  Windows computer. Moreover, I will show how to use  

• 00:01:57 it on RunPod, like this one, and I will also show  how to use it on SimplePod as well. And finally,  

• 00:02:05 I will show it how to use on Massed Compute,  our favorite cloud provider. But this is not  

• 00:02:11 all. This application also supports real-time  voice changing. So with live voice changing,  

• 00:02:17 you can speak, and as you speak, it will process  the speaking and change the voice. Furthermore,  

• 00:02:24 this application supports training your own  voice model as well. It has a full training tab,  

• 00:02:30 as you can see. It also has some other  features. So let's begin the full tutorial.

• 00:02:36 So as usual, I have prepared an amazing  post where you will find the instructions,  

• 00:02:41 information, and the installers zip file.  Let's download the latest zip file from  

• 00:02:46 here. I recommend you to read this post. You  will find a lot of important information. One  

• 00:02:53 of the most important things that you need to  do is following Windows requirements tutorial  

• 00:03:00 if you haven't yet. Then move the zip file  into the disk where you want to install. I  

• 00:03:07 am going to install into my Q drive. You can  install anywhere you want. With the installers,  

• 00:03:13 all you need to do is just run the install  and update .bat file. Make sure that your  

• 00:03:20 path doesn't have any space characters or any  special characters. Make your paths like this.

• 00:03:28 The installation will be super fast. It will  download everything automatically for you,  

• 00:03:32 including 30+ demo voices. So just patiently wait  for installation to be completed. The installation  

• 00:03:40 has been completed. Now it is downloading the  models. It will download all the necessary  

• 00:03:45 models automatically. The model downloader is a  special downloader that I have developed and I  

• 00:03:50 am using it with my all the installation  applications, all the AI applications.  

• 00:03:56 It is also verifying the hash value and it is  using 16 connections to download. Therefore,  

• 00:04:02 it is ultra fast and always accurate.  It also has full resume capability. This  

• 00:04:08 application doesn't have very big models,  but with the very big models applications,  

• 00:04:13 it is super useful. Now it is downloading the  demo voice models. There are 37 demo voices that  

• 00:04:20 we have added. But you can find hundreds of demo  voices in these links. And where you need to put  

• 00:04:28 the demo voices? Enter inside RVC Premium, enter  inside the assets, you will see there are weights.  

• 00:04:36 These are where you put the .pth files, and in the  indices, you will put the index files. By the way,  

• 00:04:44 index files are not mandatory, but they  improve the quality of the voice changing.

• 00:04:50 Okay, so the entire installation has been  completed. Now we can start the application.  

• 00:04:55 For starting the application, there are 2 options.  First one is start voice change and train web UI.  

• 00:05:02 This is what I recommend you to use. You can also  use real-time voice changing desktop application,  

• 00:05:08 which I plan to show after this. Both  of them is working. So the application  

• 00:05:13 has been started. Just wait a little bit, it  loads the model. So in the inferencing voice,  

• 00:05:19 you can see the demo voices and you can change  them. But first, let's make a default test. We  

• 00:05:25 also have options: auto separate vocals plus  music. So therefore, you can even process your  

• 00:05:31 music. It is already selecting the best one, but  you can try which one is working best for you.  

• 00:05:37 We have inference preset system which you can  make changes and save, like 1, save. It will  

• 00:05:45 save the preset entirely. Then you can set it  like this and make it like save 2, and it will  

• 00:05:50 be saved. Then you can change between them.  It will remember every settings. If you don't  

• 00:05:56 want to use any preset, go to the presets folder  and just delete it and restart your application.

• 00:06:03 Okay, so from this window, we are going  to upload an audio. You can also drag and  

• 00:06:09 drop. I already have some audio files.  For example, this 15-second one. Okay,  

• 00:06:15 let's upload it and let's listen to it. "Welcome  to the LTX 2.3 presets tutorial. In this tutorial,  

• 00:06:23 I will show you how to use LTX 2.3 models with  best quality and performance. Everything will be  

• 00:06:29 1 click to install and use." Okay, this was  my demo test voice. I am going to change it  

• 00:06:36 into Donald Trump demo voice. For example, this  one. Now, all I need to do is click convert. The  

• 00:06:45 process is extremely fast, even on low-end  GPUs. This application is very lightweight,  

• 00:06:52 and it is done. So let's listen the voice.  "Welcome to the LTX 2.3 presets tutorial.  

• 00:06:59 In this tutorial, I will show you how to  use LTX 2.3 models with best quality and  

• 00:07:05 performance. Everything will be 1 click to install  and use." As you can see, pretty decent. I mean,  

• 00:07:12 from this input to this voice is really hard  normally because they are nothing like sort of  

• 00:07:19 similar. You can play with this parameter and it  really impacts how it sounds, so I recommend you  

• 00:07:26 to try. I find that 12 is working best for this  voice, but for other voices, it may be different.

• 00:07:32 Now let's try a music. If you remember, I have  just recently published SECourses level song  

• 00:07:39 generator upgraded application. It is an  amazing application. So I have a demo song  

• 00:07:46 that I have generated with it, this one. So  let's listen to it first a little bit. Okay,  

• 00:08:06 you see the voice. I am going to use Tupac  demo voice to change it. Everything is  

• 00:08:13 default. I didn't change anything. Let's  make this like 192, save, and convert.  

• 00:08:21 Now it will convert it into the Tupac demo voice.  Let's see how it works. Okay, we got the output.  

• 00:08:29 You can also see the process logs here or  on the CMD window. So let's listen to it.

• 00:09:24 Okay, so you understand the logic. We are using  the pre-trained voices and we are able to change  

• 00:09:30 voices of the music videos or just the pure audio  speaking. Pure audio speaking is working better  

• 00:09:36 than the music videos because with the music  videos, you need to separate music and the voice  

• 00:09:42 vocals, instruments, and a lot of thing. Normally,  you need to use this page to do that. However,  

• 00:09:47 I have added it to here. This is a new stuff that  I have coded and it is automatically done for you.  

• 00:09:53 The application also supports training. If you  have a custom voice, you can train. However,  

• 00:09:58 I really don't have time to research the training,  so I am not going to show the training part.  

• 00:10:03 You can research it. There are a lot of previous  tutorials about this. There are ckpt export ONNX,  

• 00:10:09 but this is totally not necessary anymore  because it is already too fast and the  

• 00:10:13 VRAM usage is below 4 gigabytes. And  there are frequently asked questions,  

• 00:10:17 you can read it. So this is how you use  this application. You can also play with  

• 00:10:22 these parameters, but I did some research  and these are the best that I have found.

• 00:10:28 We also support batch inference, and it is  working amazing. Let me demonstrate you.  

• 00:10:34 So I have this folder for batch processing.  Click here to open the batch processing tab.  

• 00:10:40 Enter it and enter a output path like "results"  and convert. It will use your selected settings  

• 00:10:50 in the single inference and it will do the batch  processing. Every file will be processed. You can  

• 00:10:56 follow the progress on here or also on here. These  are all upgrades to the other RVC applications  

• 00:11:05 that you may find on the other repositories on  the internet. Okay, all is done. It is almost  

• 00:11:10 instant. And when I enter inside the results, I  will see the results. So let's listen few of them.

• 00:11:33 Okay, it is working pretty decent.  If you want higher quality,  

• 00:11:37 you need to find higher quality pre-trained  model like this one. So how you are accessing  

• 00:11:44 the generated files? When you click open  outputs folder, it will open the outputs  

• 00:11:48 folder. They will be automatically saved inside  here with their metadata. So you can see each  

• 00:11:54 one metadata like this. Everything, every  used settings, configuration will be saved.

• 00:12:00 Before starting cloud tutorial part, I will  also show you real-time voice change. For this,  

• 00:12:05 you need to have local computer obviously, not  on cloud. It will open this screen. Then you  

• 00:12:12 need to select your voice change .pth file.  They were downloaded inside assets/weights if  

• 00:12:18 you remember. Select it like this. Then select  the index file. The index files are saved inside  

• 00:12:24 the assets and indices, like this one. So select  the both accurate files. Then you need to select  

• 00:12:33 your input device from here, like whichever the  microphone you are using. I am using this one,  

• 00:12:39 and you can also choose the output. So now let  me show you how my voice changes live. Okay,  

• 00:12:47 now I am going to start audio conversion and  we are going to hear my voice changed. "Now  

• 00:12:56 you should hear my voice changing. This is live  changing voice. That's how you use. You can also  

• 00:13:03 change these parameters, and this is live  changing." Then don't forget to click stop.  

• 00:13:08 Then you can close it. You can also close  from here and it will be closed. That's it.

• 00:13:13 So now I will show how to install on RunPod and  then Massed Compute. For RunPod, we have RunPod  

• 00:13:19 SimplePod RVC instructions. You can use either  SimplePod or RunPod. SimplePod is cheaper. I have  

• 00:13:26 made some tutorials about it. It's exactly same  to use. RunPod is also same to use. So please use  

• 00:13:33 this link to register RunPod. If you are going to  register to SimplePod, please use this one. Let's  

• 00:13:39 see both of the prices. Okay, let's open both  of the websites. So this is SimplePod interface.  

• 00:13:47 Okay, so I am logged in into SimplePod. Let's also  log in into the RunPod. And I am in the RunPod  

• 00:13:53 interface. So please use the templates links  here. That is super important. When I click this,  

• 00:14:00 it will select the template of the SimplePod  that we have made. Then use these template links  

• 00:14:07 to get to the accurate template. From this  screen, you can use edit persistent volume,  

• 00:14:14 edit and use. Edit and use, then you can add  persistent volume. This should be workspace. You  

• 00:14:20 can make the system disk more if you are not going  to use persistent volume, like 100 gigabytes. Save  

• 00:14:26 and use. Then you need to select a GPU. You  can use any GPU, but RTX 5090 is really great,  

• 00:14:32 or RTX Pro 6000 is really great. You see the price  is $0.6 for RTX 5090 and $1 for RTX Pro 6000.

• 00:14:43 I will proceed with the RunPod because it is  same. With the RunPod, we are going to use this  

• 00:14:48 template, and the template is selected. Now, there  is one very important thing. Since we are using  

• 00:14:54 CUDA 13, you need to apply CUDA version from here.  Otherwise, it will give you older Nvidia driver,  

• 00:15:01 therefore it will not work. So select 12.9 and  13. It will work. We just need the Nvidia driver,  

• 00:15:07 therefore you can select these 2 CUDA versions.  Then select your GPU. I have tested few RTX 5090  

• 00:15:13 and recently they were all broken on RunPod.  Therefore, I am going to use RTX Pro 6000. And  

• 00:15:20 the template is selected, the accurate template.  You can also edit it and change the volume disk or  

• 00:15:27 you can also set the network volume persistent  storage, which you can set from this storage  

• 00:15:33 panel. I click here and I can set from here.  However, for this demonstration, I am going  

• 00:15:38 to use volume disk. So the volume disk will be  200 gigabytes. But you can use either of them.  

• 00:15:42 This is a new feature that you select here. And  click deploy on demand. Then you just need to wait  

• 00:15:48 it to load the storage and load the template. It  should be pretty fast usually with RTX Pro 6000.

• 00:15:57 Again, let me show you one more important  part. So it is downloading the image. Just  

• 00:16:03 wait until the connect button appear here.  You can click time to time to see. Okay,  

• 00:16:08 so it appeared. As I said, just click time  to time to see. Then click connect Jupyter  

• 00:16:15 Lab from here. Then upload the downloaded zip  file into your workspace. You know the rest  

• 00:16:22 is same actually with any other tutorials,  applications that I have. Then right-click  

• 00:16:27 and extract archive. But before proceeding  using this machine, I recommend you to open  

• 00:16:33 a terminal. Then open the RunPod SimplePod RVC  instructions. Copy this installation command,  

• 00:16:40 open a terminal, and paste it. Then it will  install it fully with downloading models and  

• 00:16:46 everything. Then we will use this command to  run it. I am fully optimizing my installers to  

• 00:16:52 be really fast on cloud machines. RunPod is very  slow, even though I do a lot of optimization, but  

• 00:16:58 Massed Compute is super fast, as you know. After  this, I will show the Massed Compute hopefully.

• 00:17:04 So the RunPod installation has been completed.  Let's start the application. For starting,  

• 00:17:09 again we return back to this file. Open a new  terminal, copy paste it, and it will start the  

• 00:17:15 application. Okay, so the application started. We  can see the Gradio live link. Currently this is  

• 00:17:21 running on RunPod. Even if I close my computer, it  will continue running. And it is running on cloud,  

• 00:17:28 not on my machine. So the usage is same.  Let's make a quick test to demonstrate you.  

• 00:17:34 Let's try this song. Okay, upload completed,  and let's just click convert. Since this is  

• 00:17:41 running with Gradio live, the file transfer may  be slow. But it will be still saved inside the  

• 00:17:48 RVC Premium inside the outputs folder, which it  will generate after the first generation. Okay,  

• 00:17:56 let's just wait a little bit. The first run  may be a little bit slow, like 30 seconds,  

• 00:18:02 but then it should get faster. You can even  process very long audios with this application.  

• 00:18:09 Hours long it is using now. And the outputs  generated. Okay, now I can see. You see the  

• 00:18:17 file generated and saved. So you can download all  of the outputs with right-click and download as  

• 00:18:22 an archive. Or you can download from here as  well, which it will appear. Let's see. Okay,  

• 00:18:31 it has appeared, so you can click this icon to  download the generated song. Or from here you  

• 00:18:37 can listen. Yeah, this is running on RunPod.  This is how you use on RunPod. This is very  

• 00:18:45 up to date of using on RunPod. And all of my  applications will work same way with RunPod.

• 00:18:53 Now I will show how to use RVC Premium on  Massed Compute. So open the Massed Compute  

• 00:18:59 instructions. Please use this link to register. I  appreciate that. I prefer Massed Compute over the  

• 00:19:06 RunPod or SimplePod for the speed, the  installation speed, the model loading,  

• 00:19:12 it is much faster. So after you registered, add  some billing. You need to also add some billing  

• 00:19:17 on the RunPod and SimplePod as well. Then go to  deploy, and you see these GPUs are available and  

• 00:19:23 these are not. I recommend you to use RTX Pro 6000  and see the price. So select the category creator,  

• 00:19:31 select the image SECourses. This is mandatory.  Then apply our coupon. Now, our coupon is  

• 00:19:38 SECourses. You see it was $1.5 per hour, and after  verify, it is $1.14. And deploy. And on RunPod,  

• 00:19:47 it is costing $1.72. But on the Massed Compute, it  will cost only $1.14. SimplePod is still cheaper  

• 00:19:57 for this GPU. However, SimplePod is not fast  as the Massed Compute. So SimplePod is like  

• 00:20:05 $1 per hour. The only disadvantage of Massed  Compute, it doesn't have a permanent storage,  

• 00:20:11 but it is much faster. You need to just wait  for initialization. After that, it is like 2  

• 00:20:17 minutes to install and run everything. It should  automatically let you know after initialized,  

• 00:20:23 but you can also refresh this page to  see whether it is initialized or not,  

• 00:20:28 like click here and deploy, then click your  running instance to refresh its status.

• 00:20:34 Okay, so our machine has been initialized. Before  I connect, I am going to copy my installer file  

• 00:20:42 into my Massed Compute shared folder. I am  going to use the ThinLinc client. It is so  

• 00:20:49 easy to install. I have shown numerous times. Just  install and go to local devices, drivers, and add  

• 00:20:57 a shared folder with read and write permission.  Okay, so copy this and paste here. Copy username,  

• 00:21:06 paste here. Copy password and paste here, and  connect. Now, continue. The installation is same  

• 00:21:15 for all of my applications, so each time you don't  need to re-watch this video. Once you've done it,  

• 00:21:22 you will understand the logic. The logic doesn't  change. So in the home, inside thin drives,  

• 00:21:29 this is the specific drives inside Massed  Compute shared folder. This is my name,  

• 00:21:34 you can give any name. Just wait a little  bit. This is not good for big file transfers,  

• 00:21:40 but for small file transfers, you can use this.  For big file transfers, you can use Hugging Face,  

• 00:21:46 Google Drive, OneDrive. You see this  has a desktop interface, like Windows.  

• 00:21:51 This is a Linux Ubuntu machine, but it has  interface. It is not like RunPod or SimplePod.

• 00:21:58 Okay, in here I will find my installer file here.  I will drag and drop it into downloads folder.  

• 00:22:04 This is mandatory. Don't forget to drag and drop  it into downloads or desktop. Then right-click  

• 00:22:09 and extract here. Enter inside the folder,  open Massed Compute instructions txt file,  

• 00:22:15 copy the install command, and open a new terminal  inside this folder. This is important. You see I  

• 00:22:22 am inside this folder. Right-click and paste and  hit enter. It will start the installation process.  

• 00:22:28 The process will be super fast compared to RunPod  or SimplePod. We can watch it live right now. The  

• 00:22:35 only thing that is taking a little bit time is  this very old fairseq library. We are using the  

• 00:22:42 GitHub repository for this. So the installation  already starting. The download speeds and  

• 00:22:49 the install once the library's dependencies are  resolved is super fast, as you can see. I am using  

• 00:22:57 UV installers in my all applications right now,  and it is super fast compared to the older pip  

• 00:23:04 installation. Still we are using pip, but it is uv  pip install, so it is making massive difference.

• 00:23:10 Okay, so the installation has been already  completed. Then we are going to use this command  

• 00:23:16 from the Massed Compute instructions read txt  file. Open a new terminal again inside this main  

• 00:23:23 folder. Right-click and paste. Then it will start  the application. You can use this application  

• 00:23:29 either inside this interface or with Gradio live.  I prefer Gradio live. You can see that it is going  

• 00:23:36 to show the GPU like this. If the machine  is broken, you won't see a GPU here. That's  

• 00:23:42 important. You can fully ignore this message.  This is just debug message, not important. So the  

• 00:23:48 Gradio live started. I will copy the link and open  it in my own computer, so I can use it as in my  

• 00:23:55 Windows. I am going to upload the audio file here.  Don't get confused, this is uploading for index  

• 00:24:02 file, but we already selected it. So I am going  to upload the audio file into here, then convert.  

• 00:24:11 It will be much faster. Okay, now it is done.  "Welcome to the LTX 2.3 presets tutorial." Okay,  

• 00:24:19 it is working amazing. I can click here to  download like this. Or I can use the generated  

• 00:24:26 output from the outputs. You see it is generated  here. I can copy them, paste them into my shared  

• 00:24:33 folder inside thin drives and download it from  my own computer. Because these are small files,  

• 00:24:40 it will work. For example, I can copy paste it  into my shared folder. Since I have a lot of files  

• 00:24:46 in my shared folder, it is taking longer, but it  should be faster in your case. Okay, for example,  

• 00:24:51 I can copy paste it here as an output folder.  Then when I go to my shared folder inside here,  

• 00:25:00 it will appear. Yes, outputs, you see  the audio file copied into my computer.

• 00:25:07 After that, I need to delete instance to not spend  any time, which I am going to do right now. So  

• 00:25:16 it will not use any of my credits. You need to  do the same with SimplePod and RunPod as well.  

• 00:25:23 You need to stop them and delete the machines.  Otherwise, they will still keep using your money.  

• 00:25:30 When you stop the pod, it will use minimal amount  of money in RunPod, but it will still use money.

• 00:25:37 If you still wonder how to use it on SimplePod,  I will show one more time. Open this link,  

• 00:25:42 it will select the template. You can use  edit and use, like adding persistent volume,  

• 00:25:47 then save and use. It will select the template  SECourses. Then select your GPU like this one,  

• 00:25:54 and run. Then it will start the machine. You will  see the Jupyter connect. The rest is totally same  

• 00:26:00 as the RunPod. Just wait for Jupyter to appear  here. On SimplePod, you can close them from this  

• 00:26:08 screen, but they will be in the my servers, all  servers. So that is how you can connect them. The  

• 00:26:14 Jupyter Lab not appeared yet, so I need to click  info from here and wait for Jupyter to appear.  

• 00:26:21 Same as in the RunPod. Okay, so the Jupyter of  the SimplePod also appeared. So the Jupyter of  

• 00:26:29 the SimplePod also arrived. Let's also run the  installation here to see the speed comparison.  

• 00:26:35 Okay, this is on the SimplePod. I want to see the  difference of the speed. Usually, the SimplePod  

• 00:26:42 is much faster. Okay, extract archive. As I said,  the RunPod and SimplePod are exactly same. That is  

• 00:26:49 how I am developing the scripts. Okay, the RunPod  installation is also about to finish. But I can  

• 00:26:56 already see that... So this is how you use on  SimplePod. You can also prefer SimplePod if you  

• 00:27:03 wish. But as you can see, the installation setup,  everything is same with my RunPod installation  

• 00:27:10 always. This is how I am keeping my scripts, the  installers organized, and how I am making them.  

• 00:27:17 That's it. I hope you have enjoyed. Thank you  so much for watching. Hopefully see you later.