YouTube Automator

YouTube Automator is a Python workflow that can generate and upload both YouTube Shorts and long-form educational videos. It picks a topic, writes a script with Gemini, generates voiceover audio, downloads royalty-free visuals and music, renders the final video with MoviePy, and uploads it to YouTube.

This README is written to help someone clone the repository on a new computer and run it successfully from scratch.

Architecture Diagram

What this project does

The pipeline supports two flows:

• Shorts flow via main.py
• Long-form flow via main_long.py

Typical output files:

• final_video.mp4 for Shorts
• final_long_video.mp4 for long-form videos
• final_long_video.srt for long-form subtitles

Features

• Topic generation for educational/fact-style content
• AI script generation using Gemini
• Voice generation with edge-tts, plus fallback to gTTS and pyttsx3
• Stock video selection from Pexels and Pixabay
• Music download and background audio support
• Automatic video rendering with MoviePy
• Automatic YouTube upload through the YouTube Data API
• Windows automation helpers via batch files and PowerShell scripts

Requirements

Before running this project on any PC, make sure you have:

• Python 3.10 or 3.11 installed
• Git installed
• A stable internet connection
• A Google account with access to YouTube uploads
• Your own API credentials for the services you want to use

Required accounts and API access

To run the full project, you should prepare these:

1. Google Gemini API key

   • Used for script generation
   • Stored in .env as GEMINI_API_KEY or GEMINI_API_KEYS

2. Pexels API key

   • Used for stock video downloads
   • Stored in .env as PEXELS_API_KEY

3. Pixabay API key

   • Used as a fallback source for stock videos
   • Stored in .env as PIXABAY_API_KEY

4. YouTube OAuth credentials

   • Download your OAuth desktop app credentials JSON from Google Cloud
   • Save it in the repository root as client_secret.json

Exact setup instructions for a new PC

1) Clone the repository

Windows / macOS / Linux:

git clone https://github.com/shiv989898/Youtube-Automator.git
cd Youtube-Automator

2) Create a virtual environment

Windows (PowerShell):

python -m venv .venv
.\.venv\Scripts\Activate.ps1

If PowerShell blocks activation, run this once in a PowerShell window:

Set-ExecutionPolicy -Scope CurrentUser RemoteSigned

Windows (Command Prompt):

python -m venv .venv
.venv\Scripts\activate.bat

macOS / Linux:

python3 -m venv .venv
source .venv/bin/activate

3) Install Python dependencies

pip install --upgrade pip
pip install -r requirements.txt

4) Create your environment file

Copy the example file:

Windows (PowerShell):

Copy-Item .env.example .env

Windows (Command Prompt):

copy .env.example .env

macOS / Linux:

cp .env.example .env

Then edit .env and add your real values.

Example:

GEMINI_API_KEY=your_gemini_key_here
PEXELS_API_KEY=your_pexels_key_here
PIXABAY_API_KEY=your_pixabay_key_here
YOUTUBE_API_KEY=optional_if_needed

5) Add your YouTube OAuth file

Place your Google OAuth desktop app credentials file in the project root as:

client_secret.json

This file is required for uploading videos to YouTube.

6) First-time YouTube authentication

On the first successful upload attempt:

• A browser window will open
• You will log in to your Google account
• You will approve YouTube upload access
• The project will save token.pickle locally for future uploads

You do not need to create token.pickle manually. It is generated automatically.

How to run the project

You can run the project either with Python directly or with the included launcher scripts.

Option A: Run directly with Python

This is the most reliable option on any PC.

Run Shorts

python main.py

Run long-form

python main_long.py

On macOS or Linux, if python does not point to the virtual environment interpreter, use:

python3 main.py
python3 main_long.py

Option B: Use the included Windows launcher files

These now work relative to the repository folder and are safe to use on any Windows PC after setup.

Run Shorts on Windows

run_shorts.bat

Run long-form on Windows

run_long.bat

What happens when you run it

Shorts flow

When you run main.py, the project will:

1. Pick a topic
2. Generate a short script
3. Clean the script for narration
4. Generate voiceover audio
5. Calculate voiceover duration
6. Download matching visuals
7. Download background music
8. Render a Short
9. Upload it to YouTube

Long-form flow

When you run main_long.py, the project will:

1. Pick a topic
2. Generate a long-form script
3. Clean the script for narration
4. Generate voiceover audio
5. Calculate voiceover duration
6. Download landscape visuals
7. Download background music
8. Render a long-form video
9. Generate subtitles
10. Upload it to YouTube

Output files

After a successful run, you should typically see files like:

• voiceover.mp3
• voiceover_long.mp3
• visual_0.mp4, visual_1.mp4, etc.
• long_visual_0.mp4, long_visual_1.mp4, etc.
• final_video.mp4
• final_long_video.mp4
• final_long_video.srt

Exact quick-start commands by operating system

Windows PowerShell

git clone https://github.com/shiv989898/Youtube-Automator.git
cd Youtube-Automator
python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install --upgrade pip
pip install -r requirements.txt
Copy-Item .env.example .env
notepad .env
# Add your API keys to .env
# Put client_secret.json in the repo root
python main.py

Windows Command Prompt

git clone https://github.com/shiv989898/Youtube-Automator.git
cd Youtube-Automator
python -m venv .venv
.venv\Scripts\activate.bat
pip install --upgrade pip
pip install -r requirements.txt
copy .env.example .env
notepad .env
python main.py

macOS / Linux

git clone https://github.com/shiv989898/Youtube-Automator.git
cd Youtube-Automator
python3 -m venv .venv
source .venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt
cp .env.example .env
nano .env
# Add your API keys
# Put client_secret.json in the repo root
python3 main.py

Running without YouTube upload

If you only want to test generation locally, you can still run the project, but the YouTube step requires valid client_secret.json and successful OAuth login.

If you want a true local-only mode later, that can be added as a feature.

Troubleshooting

ModuleNotFoundError

Your virtual environment is probably not activated, or dependencies were not installed.

Fix:

pip install -r requirements.txt

PowerShell says script execution is disabled

Run:

Set-ExecutionPolicy -Scope CurrentUser RemoteSigned

client_secret.json not found

Make sure the file exists in the repository root.

YouTube upload fails or asks for auth again

Delete token.pickle and rerun the script to trigger a fresh Google login.

Voice generation fails

The project tries edge-tts first, then gTTS, then pyttsx3 as a fallback.

Video generation feels slow

This is normal for MoviePy, especially on long-form runs.

Pexels or Pixabay returns weak results

Check that your API keys are valid and not rate-limited.

Automation options

• Windows batch launchers: run_shorts.bat, run_long.bat
• Windows looping script: loop_shorts.ps1
• Task Scheduler setup: see TASK_SCHEDULER_SETUP.md
• Cloud/CI options: EASY_DEPLOYMENT.md, GITHUB_ACTIONS_SETUP.md, FLY_DEPLOYMENT.md, RENDER_DEPLOYMENT.md, RAILWAY_SETUP.md

Project structure

• main.py: Shorts workflow entry point
• main_long.py: Long-form workflow entry point
• script_generator.py: Shorts script generation
• script_generator_long.py: Long-form script generation
• voiceover.py: Voice generation
• visuals.py: Shorts visual collection
• visuals_long.py: Long-form visual collection
• video_creator.py: Shorts video rendering
• long_video_creator.py: Long-form video rendering
• youtube_uploader.py: Shorts upload
• youtube_uploader_long.py: Long-form upload

Open source

• LICENSE
• CONTRIBUTING.md
• CODE_OF_CONDUCT.md
• SECURITY.md

Please do not commit:

• .env
• client_secret.json
• token.pickle
• generated media files
• personal credentials of any kind

License and responsible use

This project is licensed under the MIT License. See LICENSE.

You are responsible for complying with the terms of the APIs and content providers you use, including Google, Pexels, Pixabay, Microsoft, and any music or media sources referenced by the project.