Wav2vec2 Base Finetuned Ks
A speech recognition model fine-tuned on the superb dataset based on facebook/wav2vec2-base, achieving 98.34% accuracy
Downloads 13
Release Time : 3/19/2023
Model Overview
This model is a fine-tuned version of wav2vec2-base, specifically designed for speech recognition tasks, excelling in Keyword Spotting (KS) tasks
Model Features
High Accuracy
Achieves 98.34% accuracy on the evaluation set
Based on wav2vec2 Architecture
Uses facebook's wav2vec2-base as the base model
Fine-tuned on superb Dataset
Specially fine-tuned on the superb dataset
Model Capabilities
Speech Recognition
Keyword Detection
Use Cases
Voice Interaction
Smart Home Control
Used to recognize keywords in user voice commands
High accuracy in recognizing specific commands
Voice Assistant
Detects wake words or specific commands
Featured Recommended AI Models
Š 2025AIbase