S

Sepformer Wham Enhancement

Developed by speechbrain
A toolkit for speech enhancement (denoising) using the SepFormer model, pre-trained on the WHAM! dataset (8kHz sampling rate version) to remove environmental noise and reverberation.
Downloads 827
Release Time : 3/2/2022

Model Overview

This model is based on the SepFormer architecture, specifically designed for speech enhancement tasks, effectively removing environmental noise and reverberation from audio to improve speech clarity.

Model Features

Efficient speech denoising
Effectively removes environmental noise and reverberation from audio, improving speech clarity.
Transformer-based architecture
Utilizes the SepFormer architecture with self-attention mechanisms for efficient speech separation and enhancement.
Pre-trained model
Pre-trained on the WHAM! dataset, ready for direct use in speech enhancement tasks without additional training.

Model Capabilities

Speech denoising
Speech enhancement
Audio processing

Use Cases

Speech processing
Speech enhancement
Removes environmental noise and reverberation from recordings to improve speech clarity.
SI-SNR improved to 14.35 dB, PESQ reached 3.07.
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase