AIbase
EN
Home
Audio Class Finetuned
A

Audio Class Finetuned

Developed by Chemsseddine
This model is a fine-tuned audio classification model based on facebook/wav2vec2-base on the superb dataset, achieving an accuracy of 0.6578 on the evaluation set.
Audio Classification
Transformers
Open Source License:Apache-2.0#Audio Classification#wav2vec2 Fine-tuning#Low-resource Training
Downloads 20
Release Time : 4/16/2023

Model Overview

This is a fine-tuned model for audio classification, based on the wav2vec2 architecture, suitable for speech-related classification tasks.

Model Features

Based on wav2vec2 Architecture
Uses facebook's wav2vec2-base model as the foundational architecture, providing excellent audio feature extraction capabilities.
Fine-tuned on superb Dataset
Fine-tuned on the superb audio dataset, making it suitable for various audio classification tasks.
Moderate Accuracy
Achieved an accuracy of 0.6578 on the evaluation set, suitable for general audio classification applications.

Model Capabilities

Audio Classification
Speech Feature Extraction

Use Cases

Speech Processing
Speech Emotion Recognition
Can be used to identify emotional states in speech
Voice Command Classification
Can be used to recognize and classify voice commands
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