sew-mid-100k-ft-keyword-spotting Open-source Model - Free Deployment to Assist with Precise Keyword Recognition

Sew Mid 100k Ft Keyword Spotting

Developed by anton-l

This model is a fine-tuned version of asapp/sew-mid-100k on the superb dataset, primarily used for keyword spotting tasks.

Downloads 116

Release Time : 3/2/2022

Model Overview

A fine-tuned audio classification model specifically designed for keyword spotting tasks, demonstrating high accuracy on the evaluation set.

High Accuracy

Achieved a high accuracy of 97.57% on the evaluation set.

Fine-tuned Optimization

Specially fine-tuned based on the asapp/sew-mid-100k model on the superb dataset.

Audio Classification

Keyword Spotting

Voice Interaction

Voice Assistant Wake Word Detection

Used for detecting device wake words

High accuracy in recognizing specific keywords

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5999	1.0	399	0.2262	0.9635
0.4271	2.0	798	0.1230	0.9697
0.3778	3.0	1197	0.1052	0.9731
0.3227	4.0	1596	0.0975	0.9757
0.3081	5.0	1995	0.0962	0.9753

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base