W

Wav2vec2 Base Ft Keyword Spotting

Developed by anton-l
A speech keyword recognition model fine-tuned on the SUPERB dataset based on facebook/wav2vec2-base, achieving an accuracy of 98.26%
Downloads 70
Release Time : 3/2/2022

Model Overview

This model is a speech classification model for keyword recognition, fine-tuned based on the wav2vec2 architecture, excelling at detecting specific keywords from speech

Model Features

High Accuracy
Achieves an accuracy of 98.26% on the evaluation set, demonstrating excellent performance
Based on wav2vec2 Architecture
Utilizes the proven wav2vec2-base architecture as the foundational model
Efficient Fine-tuning
Targeted fine-tuning on the SUPERB dataset optimizes keyword recognition capabilities

Model Capabilities

Speech Classification
Keyword Recognition
Audio Feature Extraction

Use Cases

Smart Home
Voice Wake-up
Used for voice wake-up functionality in smart devices
High accuracy in recognizing wake words
Voice Assistants
Command Word Recognition
Identifies key command words in user voice instructions
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase