AudioEnhancerMAX Files

# AudioEnhancerMAX ### The open-source AI audio media center **Professional audio enhancement, transcription, TTS, monitoring, and edge processing.**
**16+ processing modules · Apple Silicon acceleration · local Ollama/Gemma smart mode · Android edge workers · local-first core · v3.5.1** [](LICENSE) [](https://python.org) [](https://fastapi.tiangolo.com) [](#) [](https://sourceforge.net/projects/audioenhancermax/) [Official Website](https://fabriziodegni.com/AudioEnhancerMAX/) · [SourceForge](https://sourceforge.net/projects/audioenhancermax/) · [Documentation](#documentation) · [Changelog](CHANGELOG.md) · [Contributing](CONTRIBUTING.md) AudioEnhancerMAX turns raw recordings into cleaner, production-ready audio without requiring a cloud service for the core processing pipeline.

Why It Exists

AudioEnhancerMAX is built for creators, podcasters, researchers, interviewers, and audio-heavy workflows where privacy, control, and auditability matter. It combines local DSP, AI-assisted decisions, real-time monitoring, transcription, text-to-speech, and optional Android edge workers in one FastAPI + vanilla JS application.

The default architecture is local-first: uploads, DSP output, timing history, and local model inference stay on the host machine unless you explicitly enable an integration that sends data elsewhere. You decide when to use local AI through Ollama/Gemma, when to offload DSP chunks to trusted devices on your LAN, and which output format to export.

AudioEnhancerMAX has also been recognized with the SourceForge Rising Star Award for open-source traction and community engagement.

Highlights

Quick Start

Requirements

• Python 3.10+
• FFmpeg
• macOS Apple Silicon recommended, with Linux/Windows support for core local processing
• Ollama + Gemma model optional for Smart Mode and expressive rewrite
• Android Studio/JDK 17 or 21 optional for building the Android worker

Install

git clone https://github.com/sev7enITA/AudioEnhancerMAX.git
cd AudioEnhancerMAX

python3 -m venv venv
source venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt

Run Locally

python -m uvicorn app.main:app --host 127.0.0.1 --port 8000 --reload

Open http://127.0.0.1:8000.

For LAN testing, bind explicitly and restrict browser origins:

export AEMAX_CORS_ORIGINS="http://192.168.1.20:8000,http://localhost:8000"
python -m uvicorn app.main:app --host 0.0.0.0 --port 8000

Optional AI Setup

brew install ffmpeg
brew install ollama
ollama pull gemma4:e2b

gemma4:e2b is the preferred model name when available; the app also auto-detects other local Gemma-family models exposed by Ollama. If Gemma is not available, Smart Mode falls back to deterministic heuristics and the core DSP pipeline still works.

Android Edge Worker

The native Android worker lives in android-worker. It exposes a small HTTP service on port 8877 and advertises itself to the master over UDP on trusted local networks.

cd android-worker
export JAVA_HOME="/path/to/jdk17-or-jdk21"
./gradlew assembleDebug

Prebuilt debug APKs are stored in releases.

Architecture

flowchart LR
    Browser["Browser UI"] --> API["FastAPI backend"]
    API --> DSP["DSP pipeline"]
    API --> STT["Whisper transcription"]
    API --> TTS["Edge/Kokoro TTS"]
    API --> Monitor["System monitor"]
    API --> Cluster["Cluster manager"]
    Cluster --> Android["Android edge workers"]
    DSP --> Output["Processed WAV/MP3/FLAC"]
    STT --> Transcript["TXT/SRT/VTT/JSON"]

Transparency and AI Governance

AudioEnhancerMAX should be evaluated as a local-first, AI-assisted tool, not as a black-box autonomous audio editor. Processing choices remain visible in the UI, source audio is preserved, and optional external services are documented.

For governance-sensitive workflows, keep the original audio, record the selected preset/options, review AI-generated transcripts, and document any external engine used for TTS or distribution.

Security Posture

AudioEnhancerMAX is local-first software. By default, the server is documented for 127.0.0.1, CORS is restricted to local origins, and user-controlled file IDs, download formats, versions, and preset IDs are validated before touching local storage.

Do not expose the FastAPI server directly to the public internet without adding authentication, rate limiting, storage quotas, and a reverse proxy policy.

Documentation

• Getting Started
• Feature Guide
• API Reference
• Edge Computing Setup
• Developer Onboarding
• AI Governance Notes
• Troubleshooting / FAQ
• Translation Notes

Development Checks

source venv/bin/activate
python -m compileall -q app
pip install -r requirements-dev.txt
ruff check app
bandit -r app

Project Links

• GitHub: sev7enITA/AudioEnhancerMAX
• Official website: fabriziodegni.com/AudioEnhancerMAX
• SourceForge: sourceforge.net/projects/audioenhancermax

License

AudioEnhancerMAX is released under the MIT License.