audio-search

Dealing with all unstructured data, such as reverse image search, audio search, molecular search, video analysis, question and answer systems, NLP, etc.

Official implementation of Neural Audio Fingerprint (ICASSP 2021)

Experimental Audio Search Algorithm Implementation (Shazam demo)

On-device Speech-to-Index engine powered by deep learning

This project use PANNs for audio tagging and sound event detection, and finally get audio embeddings. Then Milvus is used to search the similarity audio items.

🎵 Audio Fingerprinting Microservice | Shazam-like System A scalable microservice for audio recognition and fingerprinting, implementing Shazam-like functionality using: - Spring Boot backend - Elasticsearch for fast audio pattern matching - Docker containerization - Kafka for event streaming - MinIO for audio storage

以 Gemini Embedding 2 为核心的本地音乐智能播放工具 | A smart local music player powered by Gemini Embedding 2. [Android, Gemini API, Vector Search, Semantic Retrieval, Smart Roaming, Local Music, 音乐推荐, 向量检索, 语义搜索, 智能漫游]

Self-hosted semantic audio search engine powered by CLAP + FastAPI + Qdrant + Gradio. Search and tag sound effects or music by meaning — not just filenames.

simple tool for quickly searching and managing sound effects.

Deep Learning for Audio in CSharp and Keras

Song identification combining landmark audio fingerprinting with CNN/AST/MERT embeddings (contrastive InfoNCE), benchmarked on Jamendo with strong noise robustness.

AudioFingerprint is a production-ready, local audio fingerprinting and song identification system inspired by Shazam and Google Sound Search. It uses spectral peak extraction and combinatorial hashing to identify songs from short audio clips in milliseconds, with a clean Python + Flask architecture suitable for real-world deployments.

High-performance REAPER sound library browser with UCS, waveform audition, similarity search, and delivery tools.

Audio Deep Learning Project in Java

🗄️ Run a self-hosted Qdrant vector database using Docker with easy setup, persistent data, and integration for smooth service connections.

Multimodal semantic search across text, images, audio and video using Google's EmbeddingGemma 2, PostgreSQL + pgvector and FastAPI. One text query searches every media type in a shared 768-dimensional vector space. Runs locally with Docker Compose.

High-performance audio fingerprinting pipeline using advanced digital signal processing (DSP), spectral peak landmarks, and robust geometric hashing.

Similar file finder for images, video and audio, built with Tauri 2.

🎵 AI-powered semantic audio search engine. Find sounds by describing what you want, not file names. Built with React, FastAPI, CLAP embeddings, and Turbovec vector search.