D

Distil Wav2vec2 Adult Child Cls 37m

Developed by bookbot
An audio classification model based on the wav2vec 2.0 architecture, designed to distinguish between adult and child voices
Downloads 15
Release Time : 3/2/2022

Model Overview

This model is compressed from the wav2vec2-adult-child-cls model using knowledge distillation technology, specifically for adult/child voice classification tasks.

Model Features

Efficient Distilled Model
Compresses the original model through distillation technology, maintaining high performance while reducing parameter count
High Accuracy
Achieves 95.89% accuracy and an F1 score of 0.9624 on the test set
Lightweight
Only 37M parameters, suitable for deployment in resource-constrained environments

Model Capabilities

Audio Classification
Adult/Child Voice Recognition

Use Cases

Educational Technology
Children's Educational Applications
Identify user age groups in children's educational apps to provide age-appropriate content
Accurately recognizes children's voices, offering personalized learning experiences
Voice Analysis
Voice Data Analysis
Analyze age distribution in voice data
Quickly and accurately distinguishes between adult and child voices
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase