Open-source audio classification model of genre-recognizer-finetuned-gtzan_dset

Genre Recognizer Finetuned Gtzan Dset

Developed by pedromatias97

This is an audio classification model based on the distilhubert architecture, specifically designed for music genre recognition, achieving 89.72% accuracy after fine-tuning on the GTZAN dataset.

Audio Classification

Transformers

Open Source License:Apache-2.0 #Music genre classification #High-precision audio analysis #GTZAN fine-tuned

Downloads 183

Release Time : 9/17/2023

Model Overview

This model can classify music clips by genre, supporting recognition of 10 music genres, suitable for applications such as music classification and recommendation systems.

Model Features

High accuracy

Achieves 89.72% classification accuracy on the GTZAN test set

Lightweight

Based on the DistilHuBERT architecture, more lightweight compared to the original HuBERT model

Multi-genre recognition

Supports classification of 10 music genres

Model Capabilities

Music genre classification

Audio feature extraction

Music content analysis

Use Cases

Music recommendation system

Automatic music classification

Automatically add genre labels to songs in a music library

Improves music classification accuracy and efficiency

Music analysis

Music style trend analysis

Analyze genre distribution of music works from different periods

Helps the music industry understand market trends

🚀 music-genre-detector-finetuned-gtzan_dset

This model is a fine - tuned audio classification model based on ntu-spml/distilhubert. It is trained on the GTZAN dataset and can effectively classify music genres with high accuracy.

🚀 Quick Start

This model is a fine - tuned version of ntu-spml/distilhubert on the GTZAN dataset. It achieves the following results on the evaluation set:

Loss: 0.3892
Accuracy: 0.8972
Precision: 0.8989
Recall: 0.8972
F1: 0.8974

📚 Documentation

Model Information

Property	Details
Base Model	ntu-spml/distilhubert
Tags	generated_from_trainer
Datasets	gtzan
Metrics	accuracy, precision, recall, f1

Model Index

Name: music - genre - detector - finetuned - gtzan_dset Results:
- Task: Name: Audio Classification Type: audio - classification Dataset: Name: GTZAN Type: gtzan Metrics:
  - Name: Accuracy Type: accuracy Value: 0.8972431077694235
  - Name: Precision Type: precision Value: 0.8989153352434833
  - Name: Recall Type: recall Value: 0.8972431077694235
  - Name: F1 Type: f1 Value: 0.8974179462177999

🔧 Technical Details

Training Hyperparameters

The following hyperparameters were used during training:

learning_rate: 9e - 05
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 16
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon = 1e - 08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 7

Training Results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
2.2319	0.98	49	1.5808	0.5263	0.5682	0.5263	0.4767
1.2682	1.98	99	0.9750	0.7556	0.7524	0.7556	0.7510
0.9462	2.99	149	0.7403	0.7945	0.7964	0.7945	0.7921
0.5946	3.99	199	0.5921	0.8233	0.8281	0.8233	0.8214
0.4095	4.99	249	0.4772	0.8634	0.8663	0.8634	0.8638
0.3349	5.99	299	0.4167	0.8835	0.8866	0.8835	0.8841
0.2427	6.88	343	0.3892	0.8972	0.8989	0.8972	0.8974

Framework Versions

Transformers 4.33.1
Pytorch 1.10.2+cu111
Datasets 2.14.5
Tokenizers 0.13.3

📄 License

This model is licensed under the Apache 2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご