AIbase
EN
Home
Ast Finetuned Audioset 10 10 0.4593 Finetuning ESC 50
A

Ast Finetuned Audioset 10 10 0.4593 Finetuning ESC 50

Developed by xpariz10
This model is an audio classification model based on the AST architecture, pre-trained on the AudioSet dataset and fine-tuned on the ESC-50 dataset, achieving an accuracy of 94.64% on the evaluation set.
Audio Classification
Transformers
Open Source License:Bsd-3-clause#Audio Classification#High Accuracy#Transfer Learning
Downloads 24
Release Time : 12/7/2022

Model Overview

A deep learning model for audio classification, particularly suitable for environmental sound classification tasks.

Model Features

High Accuracy
Achieves a classification accuracy of 94.64% on the ESC-50 evaluation set.
Transformer-based Architecture
Uses the AST (Audio Spectrogram Transformer) architecture to process audio data.
Transfer Learning
Pre-trained on the AudioSet dataset and fine-tuned on ESC-50.

Model Capabilities

Environmental Sound Classification
Audio Feature Extraction
Sound Event Detection

Use Cases

Environmental Monitoring
Urban Sound Classification
Identify and classify various sounds in urban environments.
Can accurately recognize multiple environmental sounds such as traffic, construction, and nature.
Smart Home
Home Anomaly Sound Detection
Detect abnormal sounds in home environments, such as glass breaking or alarms.
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
EnglishįŽ€äŊ“中文įšéĢ”中文ãĢãģんご
Š 2025AIbase