gtzan

Pytorch and TensorFlow data loaders for several audio datasets

Hackville 2025 example of CreateML & CoreML for iOS platforms

Clasificación de los géneros musicales utilizando técnicas de aprendizaje profundo (CNN y LSTM) en el conjunto de datos GTZAN. Classification of musical genres using Deep Learning techniques (CNN and LSTM) on the GTZAN dataset.

Music genre classification on the GTZAN dataset

GTZAN - Music Genre Classification

A convolutional neural network pipeline for music genre classification. Raw audio files are converted to mel-spectrogram images, which are then used to train and compare three CNN architectures. Includes training curves, confusion matrices, and automatic best-model saving.

A full-stack multimodal music agent for audio analysis, playlist planning, and LLM-based music recommendation.

A ML approach to find genre of a given song file and to recommend similar songs in the same dataset.

🎵 Music Genre Classification with 90% accuracy on GTZAN dataset using CNNs and mel-spectrograms. State-of-the-art deep learning approach with complete documentation.

Music Genre Classification and Recommendation

Audio genre classification using KNN on GTZAN dataset

10-genre GTZAN classification from mean MFCCs with an MLP, evaluated with repeated cross-validation (61% vs 10% chance), duplicates handled, SVM/logistic baselines.

CNN that classifies music into 10 genres from mel spectrograms (GTZAN dataset), using audio data augmentation, batch normalization, and learning rate scheduling. Built with TensorFlow/Keras and librosa.

Real-time music-genre classification: spectrogram CNN, ONNX-optimised, served as a streaming/chunked classifier with PyTorch-vs-ONNX benchmarks. Track-aware GTZAN eval.

FastAPI service for music genre classification (GTZAN, 10 classes) using a CNN with Mel spectrograms. 78% test accuracy, SpecAugment data augmentation, fully offline inference.

Music Genre Classification using Logistic Regression

Comparative study of six neural network architectures for music genre classification using the GTZAN dataset

端到端音乐情感分析：37 维声学特征工程 → 传统机器学习分类 → Gradio 交互演示。基于 GTZAN，含弱监督局限声明。

Music genre classification with honest, leakage-audited evaluation — SHAP explanations and MMR playlist recommendation over GTZAN. Reports 89.7% / 79.3% / 51.4% across three split protocols instead of just the flattering one.