A collection of Python scripts for optimizing media assets for Unity projects, especially Unity WebGL builds.

Resizes images to have dimensions divisible by 4 (required for efficient texture compression) with a 4K maximum limit.

Features:

• Maintains aspect ratio
• Ensures dimensions are divisible by 4
• 4K maximum limit (3840px)
• Supports PNG, JPG, JPEG
• Recursive directory processing
• In-place resizing with quality preservation
• Optional PNG vertical flipping for Unity texture coordinate conversion

Usage:

# Resize all images in a directory (recursive)
python resize_images.py /path/to/images/

# Resize single image
python resize_images.py image.png

# Custom max dimension and quality
python resize_images.py /path/to/images/ --max-dimension 2048 --quality 90

# Process only current directory (no subdirectories)
python resize_images.py /path/to/images/ --no-recursive

# Flip PNG images vertically (useful for Unity texture coordinates)
python resize_images.py /path/to/images/ --flip-png

Compresses PNG and JPEG images to reduce file size while preserving visual quality.

Features:

• Smart compression (only files above size threshold)
• Format-specific optimization
• RGBA to RGB conversion for JPEG
• Progressive JPEG encoding
• Maximum PNG compression
• Default 2MB minimum file size

Usage:

# Compress all images in directory
python compress_images.py /path/to/images/

# Compress single image
python compress_images.py image.png

# Custom quality and size threshold
python compress_images.py /path/to/images/ --quality 90 --min-size 1.0

# Process only current directory
python compress_images.py /path/to/images/ --no-recursive

Converts audio files to Unity WebGL compatible formats (OGG, MP4/AAC).

Features:

• Multiple input formats (MP3, WAV, OGG, M4A, AAC, FLAC)
• Unity WebGL optimized output (OGG recommended)
• Customizable quality/bitrate
• Batch conversion
• File size reporting

Usage:

# Convert all audio to OGG (recommended for Unity WebGL)
python convert_audio.py /path/to/audio/

# Convert to MP4/AAC format
python convert_audio.py /path/to/audio/ --format mp4

# High quality conversion
python convert_audio.py /path/to/audio/ --quality 256k

# Convert to specific output directory
python convert_audio.py /path/to/audio/ --output-dir converted/

# Convert single file
python convert_audio.py audio.mp3

Generates SRT subtitle files from audio using OpenAI Whisper speech recognition.

Features:

• Automatic speech recognition using Whisper AI
• Multiple model sizes (tiny, base, small, medium, large)
• Language detection or specific language selection
• Batch processing for multiple files
• Progress tracking and summary statistics
• Properly formatted SRT with timestamps
• Skip existing files to avoid reprocessing

Usage:

# Generate SRT for all MP3 files in directory
python generate_srt_files.py /path/to/audio/

# Use specific Whisper model (base is default)
python generate_srt_files.py /path/to/audio/ --model medium

# Specify language (auto-detect by default)
python generate_srt_files.py /path/to/audio/ --language en

# Process single file
python generate_srt_files.py audio.mp3

# Output to different directory
python generate_srt_files.py /path/to/audio/ --output subtitles/

All scripts require Python 3.6+ and the following packages:

# Basic dependencies for image and audio processing
pip install Pillow pydub

# For SRT generation (optional, only if using generate_srt_files.py)
pip install torch torchvision torchaudio  # PyTorch
pip install git+https://github.com/openai/whisper.git  # OpenAI Whisper

For audio conversion and SRT generation, FFmpeg is also required:

• macOS: brew install ffmpeg
• Ubuntu/Debian: sudo apt install ffmpeg
• Windows: Download from ffmpeg.org

# Clone repository
git clone [email protected]:Cosmin-B/unity-media-tools.git
cd unity-media-tools

# Install basic dependencies
pip install Pillow pydub

# Optional: Install Whisper for SRT generation
pip install torch torchvision torchaudio
pip install git+https://github.com/openai/whisper.git

# Make scripts executable (optional)
chmod +x *.py

• Dimensions divisible by 4: Required for efficient texture compression (DXT, ASTC)
• 4K limit: Prevents memory issues on lower-end devices
• Compression: Reduces download size and loading times

• OGG format: Best compression and WebGL compatibility
• Avoid MP3: Can cause compatibility issues in Unity WebGL
• MP4/AAC: Good alternative to OGG
• Bitrate: 192k is good balance of quality/size, 128k for smaller files

1. Resize images to ensure proper dimensions:

   python resize_images.py Assets/Textures/

2. Compress images to reduce file size:

   python compress_images.py Assets/Textures/

3. Convert audio to WebGL-compatible formats:

   python convert_audio.py Assets/Audio/ --format ogg

4. Generate subtitles from audio narration (optional):

   python generate_srt_files.py Assets/Audio/ --model base

MIT License - feel free to use in your projects!

Pull requests welcome! Please ensure scripts remain generic and don't include project-specific code.