AIbase
EN
Home
Wav2vec2 Base POSITIVE NEGATIVE ONLY BALANCED CLASSES
W

Wav2vec2 Base POSITIVE NEGATIVE ONLY BALANCED CLASSES

Developed by aherzberg
A fine-tuned speech processing model based on facebook/wav2vec2-base, focusing on balanced positive-negative class classification tasks
Audio Classification
Transformers
Open Source License:Apache-2.0#Voice feature extraction#High-accuracy classification#Audio signal processing
Downloads 17
Release Time : 11/23/2022

Model Overview

This model is a fine-tuned version of wav2vec2-base, primarily used for speech-related binary classification tasks, achieving an accuracy of 88.22% on the evaluation set

Model Features

Balanced positive-negative class processing
Optimized specifically for class imbalance issues in classification tasks
High accuracy
Achieves a classification accuracy of 88.22% on the evaluation set
Based on wav2vec2 architecture
Utilizes the mature wav2vec2-base model for fine-tuning, with excellent speech feature extraction capabilities

Model Capabilities

Speech classification
Binary classification task processing
Imbalanced data classification

Use Cases

Speech analysis
Speech emotion classification
Determines the emotional tendency of speech segments
Accuracy 88.22%
Voice command recognition
Identifies whether voice commands belong to specific categories
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