spectrogram-analysis 话题在 GitHub 上没有填写简介，这里按 Star 数汇总最相关的热门仓库。
🔇 A production-grade deep learning system for real-time drone/UAV detection through acoustic signature analysis. Converts raw audio to Mel-Spectrograms and classifies using a custom CNN. Features auto-dataset ingestion, defense-optimized metrics (high recall), early stopping, model checkpointing, and a ready-to-use inference API.
Advanced radar-based classification system for detecting and distinguishing UAVs, birds, and RC aircraft using SVMD signal decomposition and deep learning feature extraction.
A tool designed to manage annotations for bioacoustics.
The data we are going to use consists of 2D spectrograms of deep space radio signals collected by the Allen Telescope Array at the SETI Institute. We will treat the spectrograms as images to train an image classification model to classify the signals into one of four classes.
Invertible spectrograms for PyTorch: audio → perceptual spectrogram (mel/Bark/ERB) → audio. Lightweight, interpretable, CPU-fast.
AI-Powered Environmental Monitoring System for detecting deforestation, poaching, and environmental hazards
Analyse automatique de chants d'oiseaux. Filtrage FIR, STFT, VAD spectral flux & Spectrogrammes
WinUI 3 realtime spectrogram analyzer, .NET 10 C# - the full 93K surface replaced every 25ms by a per-row horizontal shift. Zero-copy UseDataAtLocation Z buffer, FastCopyFrom X/Y, ReuseDataX/Y, GPU compute shader. PesgoWinUI, Direct3D.
Spectrogram Slicer & Windowizer is a modern, interactive GUI tool built using CustomTkinter for efficiently handling batches of 2D spectrogram data (stored as .npy NumPy arrays).
Repo for BickGraphing Code
See what audio processing actually does — interactive STFT explorer with an A/B spectrogram difference view.
This project explores various approaches for audio classification using neural networks with TensorFlow and Keras. The notebook demonstrates the complete process from data loading and preprocessing to model building, training, evaluation, and inference.
A classifier model built to classify music snippets into 10 pre-defined genres with CNNs. Uses spectral and MFCC data to select features.
RF-based drone model classification on the DroneDetect dataset. Features a lossless Parquet/DuckDB pipeline, spectral EDA, and an interpretable normalized Welch PSD + LDA classifier achieving 97% segment / 100% recording accuracy, benchmarked against a Spectrogram CNN.
Audio intelligence platform for multi-method steganography, blind steganalysis and forensic signal analysis.
Convolutional neural network for elephant rumble audio denoising