AIbase
EN
Home
Distil Ast Audioset Finetuned Cry
D

Distil Ast Audioset Finetuned Cry

Developed by jstoone
An audio classification model fine-tuned on the DonateACry dataset based on bookbot/distil-ast-audioset, designed for identifying infant cries
Audio Classification
Transformers
Open Source License:Apache-2.0#Infant cry classification#Lightweight audio model#AST distillation architecture
Downloads 76
Release Time : 11/5/2023

Model Overview

This is an audio classification model specifically fine-tuned for infant cry recognition tasks, achieving 63.6% accuracy on the evaluation set

Model Features

Lightweight audio classification
Based on DistilAST architecture, reducing model size while maintaining good performance
Optimized for infant cries
Fine-tuned on DonateACry dataset, focusing on infant cry recognition tasks
Fast training convergence
Achieves optimal performance within 10 training epochs

Model Capabilities

Audio classification
Infant cry recognition
Sound event detection

Use Cases

Infant care
Infant cry monitoring
For smart infant monitoring devices to detect baby cries in real-time
Recognition accuracy 63.6%
Medical assistance
Newborn care
Automatic cry monitoring system for hospital neonatal wards
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