🎵 AudioManager v4.0.0 — Phoenix

Professional Audio Transcription & Processing

Transcribe audio files locally with WhisperX — offline, secure, and blazing fast.

Download Now · User Manual · Report Issue

---

📸 Screenshots

Processing Queue & Controls	Active Transcription	Batch Report Complete

---

✨ Features

Feature	Description
🎙️ Batch Processing	Process hundreds of files with adaptive concurrency that scales to your hardware.
📝 WhisperX Transcription	State-of-the-art speech recognition with forced alignment for word-level timestamps.
👥 Speaker Diarization	Identify who spoke when with speaker labels and summaries.
⏱️ Time Estimation	Learns from past runs to predict completion times for each file.
🌐 REST API	Full REST API for headless automation and integration with other tools.
📊 Performance Reporting	Detailed stage-by-stage timing reports for every file processed.
🎨 Dark Mode	Full dark mode support for comfortable late-night work.
🔒 100% Offline	Your audio and transcripts never leave your machine.
🎵 Sound Recorder	Record audio directly into the app and add it to the queue.
🏷️ ID3 Tagging	Generate sidecar .meta files with title, artist, and metadata.
📁 Watch Folder	Automatically add files dropped into a watched folder.
📋 SRT / TXT / DOCX	Export transcripts in multiple formats, including Word-compatible .docx.
⚡ GPU Acceleration	Up to 3x faster transcription with NVIDIA GPUs.
🎬 Streaming	Automatic chunking for files >100MB to prevent memory issues.
🚀 Setup Wizard	Guided setup for dependencies, Python version validation, and model selection.

---

🚀 Download

Secure Windows Installers

Platform	Download Link	Size
Windows (EXE)	⬇ Download AudioManager-Setup.exe	~197 MB
Windows (MSI)	⬇ Download AudioManager-Installer.msi	~197 MB

🔒 Security Note: We distribute native Windows installers (.exe/.msi) instead of a .jar file to protect our licensing logic and provide a seamless installation experience. This is a deliberate security decision.

📦 System Requirements

Requirement	Details
OS	Windows 10 or later (64-bit)
Java	Bundled with the installer (no separate installation required)
RAM	4 GB minimum (8 GB recommended)
Storage	500 MB for the app + space for models and audio files
Python	3.10, 3.11, or 3.12 (3.13+ NOT supported)
GPU	NVIDIA GPU with 4+ GB VRAM (optional, for GPU acceleration)
Dependencies	FFmpeg and FFprobe (required) — see Troubleshooting Guide

⚠️ Python 3.13+ is not supported by WhisperX. See Installation → Step 1 for details.

---

📖 Quick Start

1. Download the installer for your platform from the Download section.
2. Install the app by running the installer.
3. Launch AudioManager from your Start Menu.
4. Run the Setup Wizard: Help → Run Setup Wizard (or press Ctrl+Shift+S) to:
   • Check your Python version
   • Verify FFmpeg and WhisperX installation
   • Select and verify models
5. Check dependencies: Help → Check Dependencies (or press F5).
6. Add files to the queue using the file browser or drag-and-drop.
7. Configure transcription settings (Tools → Transcription Settings).
8. Press Start and watch your files get transcribed!

---

🛠️ Installation & Setup

Step 1: Install Python (Required for Transcription)

⚠️ IMPORTANT: Python 3.10, 3.11, or 3.12 is REQUIRED. Python 3.13+ is NOT supported.

Check if Python is installed:

python --version

If Python is not installed or you have Python 3.13+:

1. Download Python 3.12 from python.org
2. Run the installer
3. IMPORTANT: Check "Add Python to PATH" during installation
4. Verify installation:

   python --version
   # Should show: Python 3.12.x

Step 2: Install FFmpeg (Required)

AudioManager relies on FFmpeg for audio processing. Install it from:

• Download: https://ffmpeg.org/download.html
• Windows: Place the ffmpeg.exe and ffprobe.exe in C:\AI\ffmpeg\bin\ or add them to your system PATH.

Verify FFmpeg installation:

ffmpeg -version
ffprobe -version

Step 3: Install WhisperX (Required for Transcription)

WhisperX is the transcription engine. Install it in a Python virtual environment:

# Create a virtual environment
python -m venv whisperx_env

# Activate it
# On Windows:
whisperx_env\Scripts\activate
# On macOS/Linux:
source whisperx_env/bin/activate

# Install WhisperX
pip install whisperx

If you get an error, try:

# Upgrade pip first
pip install --upgrade pip

# Then install WhisperX
pip install whisperx

💡 Tip: If you have multiple Python versions, use the full path:

C:\Python312\python.exe -m venv whisperx_env

⚠️ Windows Users Note: If you encounter an 'activate' is not recognized error, or if typing python opens a >>> shell instead of showing the version, your Windows "App Execution Aliases" are interfering. To fix this:

1. Open Windows Settings and search for "Manage app execution aliases".
2. Scroll down to "App Installer" (python.exe and python3.exe) and toggle them OFF.
3. If creating the environment still fails with "Access Denied", open your Command Prompt as Administrator and run python -m venv whisperx_env.

Step 4: Install TorchCodec (Recommended for Windows)

TorchCodec improves audio loading performance on Windows:

# Activate your environment first, then:
pip install torchcodec==0.7.0

⚠️ Note: If you see DLL errors after installing TorchCodec, you may need to copy FFmpeg DLLs to the TorchCodec folder. See the Troubleshooting Guide for details.

Step 5: Install Whisper Models

AudioManager does not download models automatically. To install a model:

huggingface-cli download Systran/faster-whisper-<model-name>

Replace <model-name> with tiny, base, small, medium, or large-v3.

Step 6: (Optional) Install CUDA for GPU Acceleration

If you have an NVIDIA GPU:

1. Install NVIDIA drivers from nvidia.com
2. Install CUDA Toolkit 12.x from NVIDIA CUDA
3. Verify with: nvidia-smi

---

📚 Documentation

Document	Link	Description
User Manual	USER_MANUAL.md	Complete user guide
Troubleshooting Guide	TROUBLESHOOTING.md	Common issues and solutions
API Documentation	API.md	REST API reference
Changelog	CHANGELOG.md	Version history
License	LICENSE	Apache 2.0

---

⌨️ Keyboard Shortcuts

Shortcut	Action
Ctrl+Shift+D	Toggle Dark Mode
F5	Check Dependencies
Ctrl+Shift+W	Toggle Folder Watch
Ctrl+Shift+P	Performance Report
Ctrl+Shift+S	Run Setup Wizard (NEW)
Ctrl+B	Batch Settings
Ctrl+Q	Exit
Ctrl+Shift+C	Clear Session Data
Ctrl+Z	Undo
Ctrl+Y / Ctrl+Shift+Z	Redo
F1	Troubleshooting Guide

---

🎯 Supported Formats

Audio Formats

MP3, WAV, FLAC, OGG, M4A, WMA, AAC, OPUS, ALAC, AIFF, AMR, AC3

Output Formats

SRT, TXT, DOCX, HTML, JSON

---

⚡ Performance

Model Comparison

Model	Speed	Accuracy	Memory
Tiny	⚡ Very Fast	🟡 Basic	39 MB
Base	⚡ Fast	🟢 Good	74 MB
Small	🟢 Fast	🔵 Better	244 MB
Medium	🟡 Moderate	🟣 High	769 MB
Large	🔴 Slow	⭐ Best	1.5 GB

Performance Tips

Tip	Benefit
Use GPU	2-3x faster transcription
Smaller model	Faster processing, less memory
Reduce parallel files	Lower memory usage
Enable streaming	Better memory efficiency for large files
Disable diarization	Faster on CPU

---

📊 Benchmarks & Research

This project is backed by a reproducible ablation study (207 runs) published on Zenodo:

• Dataset & Paper: AburimeSoundManager WhisperX Ablation Study
• Analysis Code: aburime-som-whisperx-ablation
• Covers: concurrency modes, ablation configurations, batch sizes, clip durations, and persistent runtime estimation accuracy.

---

🤝 Contributing

We welcome contributions! Here's how you can help:

1. Fork the repository and create your feature branch.
2. Write clean, well-documented code.
3. Add tests for new features.
4. Submit a pull request with a clear description of your changes.

Please read our CONTRIBUTING.md for more details.

Development Setup

# Clone the repository
git clone https://github.com/gregjava/Aburime-Sound-Manager.git
cd Aburime-Sound-Manager

# Build the project
ant build

# Run tests
ant test-all

# Generate test report
ant test-report

# Create distribution
ant dist

Testing

# Run unit tests
ant test

# Run integration tests
ant test-integration

# Run all tests with coverage
ant test-with-coverage

# Run a single test class
ant test-single -Dtest.class=audiomanager.util.TimeLeftEstimatorTest

Build

# Build JAR
ant jar

# Build with GPU support
ant gpu-build

# Build signed distribution
ant release

# Build distribution package
ant dist

---

🐛 Reporting Issues

If you encounter a problem, please open an issue on GitHub with:

• A clear description of the issue.
• Steps to reproduce it.
• Your system details (OS, RAM, Java version).
• Your Python version (python --version).
• The error message from the log panel.
• Output of Help → Check Dependencies.
• Output of Help → Run Setup Wizard (if applicable).

---

📄 License

This project is licensed under the Apache License 2.0 — see the LICENSE file for details.

Third-Party Components

Component	License
FFmpeg	LGPL
WhisperX	MIT
JavaFX	GPL with Classpath Exception
Apache Log4j/SLF4J	Apache 2.0
Gson	Apache 2.0

---

🙏 Acknowledgments

• WhisperX — For providing state-of-the-art transcription with forced alignment.
• FFmpeg — For handling all audio processing and conversion.
• JavaFX — For the beautiful cross-platform UI framework.
• OpenAI Whisper — For the foundational transcription model.
• HuggingFace — For model hosting and the pyannote diarization model.

---

📊 Project Status

Metric	Status
Build	✅ Passing
Tests	✅ 25+ test classes
Documentation	✅ Complete
Code Signing	✅ Windows
License	✅ Apache 2.0
GPU Support	✅ CUDA 12.x
Setup Wizard	✅ Complete

---

📞 Support

Channel	Link
Website	https://poseidon.org.uk/Aburime-Sound-Manager-v4.0/
GitHub	https://github.com/gregjava/Aburime-Sound-Manager
Issues	https://github.com/gregjava/Aburime-Sound-Manager/issues
Email	support@audiomanager.app

---

💼 Hire Me

I build production audio/ASR tooling in Java and Python — JavaFX desktop apps, WhisperX pipelines, and REST APIs.

Available for freelance and full-time remote work. 📧 tefdrive01@gmail.com

---

Made with ❤️ by Greg Java

🌐 Website · 🐙 GitHub · 🐛 Issues