DSRE / Deep Sound Resolution Enhancer

Overview

DSRE is a high-performance audio enhancement tool that batch-converts any supported audio file into high-resolution (Hi-Res) audio. Inspired by Sony DSEE HX, it uses a non-deep-learning frequency enhancement algorithm, so it can process large collections of files quickly without demanding hardware.

Key Features

• Batch Processing - convert many audio files in one pass.
• Multiple Formats - supports WAV, MP3, FLAC, M4A, OGG, AIFF, AAC, WMA, MKA, and more.
• Preserves Cover Art & Metadata - propagates tags and embedded art into the enhanced output.
• Flexible Parameters - tune modulation count, decay, filter order, and high-pass cutoffs.
• Fast & Stable - no heavy neural networks; everything runs on lightweight DSP.

---

Installation & Usage

Download the latest release from the releases page, extract it, and launch the executable. Add one or more audio files, choose an output directory, adjust parameters if needed, then click Start Processing.

---

Building with PyInstaller

You can recreate the standalone desktop build with PyInstaller. The steps below assume you are inside the project root and have Python 3.11+ available.

1. Create and activate a virtual environment (optional but recommended):

   python -m venv .venv
   source .venv/bin/activate   # Windows: .venv\\Scripts\\activate
   python -m pip install --upgrade pip

2. Install the Python dependencies:

   pip install -r requirements.txt

3. Download a static FFmpeg build for your platform and drop the ffmpeg executable (add ffprobe only if you need it) into an ffmpeg/ folder that sits next to DSRE.py. The app dynamically adds this folder to PATH both at runtime and when bundled, so the binary just needs to live inside ffmpeg/.
4. Run PyInstaller. The command below produces a single-file GUI build, ships the icon, and vendors the FFmpeg folder so it is unpacked alongside the executable at runtime:

   pyinstaller \
       --noconfirm --clean --onefile --windowed \
       --name DSRE --icon logo.ico \
       --add-data "logo.ico;." \
       --add-data "ffmpeg/*;ffmpeg" \
       DSRE.py

   (On macOS/Linux replace the semicolons in --add-data with colons.)
5. The resulting build lives in dist/DSRE/DSRE (macOS/Linux) or dist/DSRE.exe (Windows). You can zip that file (plus the unpacked ffmpeg data directory that PyInstaller drops next to it on first run) and distribute it.

If you need to tweak the bundle later (e.g., add more data files or hidden imports), re-run the same command after editing it or generate a DSRE.spec file with pyi-makespec and check it into version control.

---

Parameters

Parameter	Default	Description
Modulation count (m)	8	Number of enhancement repetitions; higher values push more high-frequency detail
Decay	1.25	Controls attenuation of boosted high frequencies
Pre-processing high-pass cutoff	3000 Hz	High-pass filter applied before enhancement
Post-processing high-pass cutoff	16000 Hz	High-pass filter applied after enhancement
Filter order	11	Order of the Butterworth high-pass filters
Target sampling rate	96000 Hz	Sample rate of the exported audio
Output format	ALAC/FLAC	Choose the Hi-Res output container/codec

---