H

Hubert Large Superb Er

Developed by superb
An emotion recognition model based on Hubert-Large pre-trained model for predicting emotion categories in speech
Downloads 10.24k
Release Time : 3/2/2022

Model Overview

This model is a fine-tuned emotion recognition model based on hubert-large-ll60k pre-trained model, primarily used for speech emotion classification tasks, supporting 16kHz sampled speech input.

Model Features

Based on Hubert-Large Pre-training
Uses hubert-large-ll60k as the base model, which is pre-trained on 16kHz sampled speech audio
Standard Evaluation Protocol
Follows the standard evaluation protocol of IEMOCAP dataset, retaining four emotion categories with similar data quantities
High Accuracy
Achieves an accuracy rate of 67.62% on session1 of IEMOCAP dataset

Model Capabilities

Speech Emotion Recognition
Audio Classification
Multi-emotion Category Prediction

Use Cases

Sentiment Analysis
Customer Service Emotion Analysis
Analyze customer emotional states in call center conversations
Can recognize basic emotional states such as happiness, neutrality
Mental Health Monitoring
Track user mood changes through voice analysis
Can assist in diagnosis of psychological disorders like depression
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase