AddictedCS/soundfingerprinting

Open source audio fingerprinting in .NET. An efficient algorithm for acoustic fingerprinting written purely in C#.

What it solves

This project provides a way to identify audio and video content by creating unique "fingerprints" of media files. It allows users to recognize unknown audio or video samples by comparing them against a database of known fingerprints, which is useful for detecting duplicates, identifying songs, or detecting ads in radio streams.

How it works

The framework extracts acoustic and video fingerprints (hashes) from media files and stores them in a configurable datastore (such as an in-memory store or the Emy persistent storage). When a query is performed, the the framework analyzes a sample of the audio or video and searches the datastore for similar sub-fingerprints to find the best match.

It also includes a spectral-profile path bridging mechanism to handle regions where hashes don't match due to broadband noise or silence, ensuring more robust recognition across different recordings.

Who it’s for

Developers, researchers, and companies specializing in digital signal processing, data mining, and audio/video recognition.

Highlights

  • Multimodal Recognition: Supports both audio and video fingerprinting, allowing for the same-time identification of both.
  • Robustness: Handles pitch and tempo changes (vinyl-style) and uses spectral-profile bridging to overcome broadband noise and silence.
  • Flexible Storage: Offers an in-memory storage option for small projects and supports Emy for persistent, large-scale storage.
  • C#/.NET Framework: Built as a NuGet package for easy integration into .NET applications.

Related

  • Project
    acoustid/chromaprint

    An audio fingerprinting library that generates compact identifiers to detect near-identical audio files and streams for high-performance searching.

  • Project
    ZFTurbo/Music-Source-Separation-Training

    An open‑source Python framework for training, evaluating, and running inference with many state‑of‑the‑art music source‑separation models. It provides config‑driven training (single‑GPU or multi‑GPU), LoRA fine‑tuning, a CLI, a small GUI, and scripts for ensembling and converting models to ONNX/TensorRT.

  • Project
    NVIDIA-AI-Blueprints/video-search-and-summarization

    NVIDIA’s Video Search & Summarization (VSS) blueprint is a GPU‑accelerated reference stack that lets you search, query, and summarise live or recorded video with natural‑language. It combines vision‑language models, LLMs, and NVIDIA NIM micro‑services, offering real‑time analytics, alert verification, and long‑video summarisation via modular Python/Node/Next.js services and Docker‑Compose or cloud‑launchable deployment.

  • Project
    hank-ai/darknet

    An open-source neural network framework written in C/C++ and CUDA that powers the YOLO real-time object detection system.

  • Project
    MTG/essentia

    An open-source C++ and Python library for audio analysis and music information retrieval, providing a wide array of signal processing algorithms and music descriptors.