🎧 SplitBySilence

A lightweight C# .NET 9 console application that automatically splits a long MP3 file into individual songs using silence detection powered by FFmpeg.

It finds silent gaps between songs, extracts each track losslessly (no re-encoding), and saves them as separate MP3 files with a CSV manifest.
Ideal for DJ mixes, continuous albums, or any long recordings.

---

🚀 Features

• 🎵 Detects silence gaps between songs automatically
• ⚡ Splits MP3s without quality loss (stream-copy mode)
• 📊 Generates a CSV manifest with timestamps
• 🧠 Adjustable parameters (--silenceDb, --minSilence, --minTrack, --pad)
• 💻 Cross-platform — runs on Windows, macOS, or Linux (requires FFmpeg)
• 🪶 Written in pure .NET 9 C#, no external NuGet dependencies

---

🛠️ Requirements

Component	Purpose	Installation
.NET 9 SDK	Run/build the console app	Download .NET 9
FFmpeg	Detect & cut silence	Download FFmpeg or winget install Gyan.FFmpeg

After installing FFmpeg, verify it’s on your PATH:

ffmpeg -version
ffprobe -version

---

🧩 Installation

Clone this repo:

git clone https://github.com/<your-username>/SplitBySilence.git
cd SplitBySilence

Build it:

dotnet build

---

▶️ Usage

Quick start

dotnet run --project SplitBySilence.csproj -- --input "C:\music\long_mix.mp3"

This will:

• Detect silences (default: quieter than -35 dB for ≥ 1.5 s)
• Split tracks
• Save them in a folder next to the input, e.g. long_mix_split/
• Generate long_mix_tracks.csv with segment info

---

Advanced options

Option	Default	Description
--input	(required)	Path to input MP3 file
--outdir	auto	Output directory
--silenceDb	-35	Silence threshold (lower = more sensitive)
--minSilence	1.5	Minimum silence duration (seconds)
--minTrack	60	Minimum valid track length (seconds)
--pad	0.0	Add/subtract seconds to cut edges
--ffmpeg	ffmpeg	Custom path to ffmpeg
--ffprobe	ffprobe	Custom path to ffprobe
--dry-run	—	Detect & show planned splits (no files written)
--verbose	—	Print detailed detection logs

Example:

dotnet run --project SplitBySilence.csproj -- \
  --input "C:\mixes\session.mp3" \
  --silenceDb -40 \
  --minSilence 2.0 \
  --minTrack 75 \
  --pad 0.15

---

📂 Output structure

📁 long_mix_split/
 ├── long_mix_track_01.mp3
 ├── long_mix_track_02.mp3
 ├── long_mix_track_03.mp3
 ├── ...
 └── long_mix_tracks.csv

The CSV manifest contains:

Index,Start,End,Duration,File
1,0.0,243.5,4:03,long_mix_track_01.mp3
2,243.5,496.2,4:13,long_mix_track_02.mp3
...

---

🧠 How it works

1. FFmpeg “silencedetect” filter analyzes the waveform:

   silencedetect=noise=-35dB:d=1.5
   

2. Logs silence start/end points → parsed by C#

3. Segments are computed and exported via ffmpeg -ss -to -c copy (no re-encode)

This ensures frame-accurate, lossless MP3 cuts.

---

🖥️ Run from VS Code

1. Install Visual Studio Code

2. Open folder:

   code .

3. Press F5 to build & run (launch config included in .vscode/launch.json)

---

🧰 Build & publish as a standalone EXE

You can package it for Windows:

dotnet publish -c Release -r win-x64 --self-contained true -p:PublishSingleFile=true

Output:

bin\Release\net9.0\win-x64\publish\SplitBySilence.exe

Then you can run:

SplitBySilence.exe --input "C:\music\album.mp3"

---

💡 Tips

• Lower --silenceDb (e.g. -40) if silences aren’t detected.
• Increase --minSilence (e.g. 2.0) to avoid false splits.
• Use --dry-run --verbose to preview detection logs before cutting.
• Works with WAV, FLAC, or other formats too — just change the file extension.

---

🧾 Example screenshot

(Replace this with your own later)

---

🧑‍💻 Author

Majid — Software Developer & Solution Architect 🔗 LinkedIn • 💻 [.NET / Blazor / FFmpeg enthusiast]

---

🪪 License

This project is licensed under the MIT License — free to use, modify, and distribute.

---

⭐ Support

If you find this helpful, please ⭐ the repo — it helps others discover it!