# Auditory_model_based_MMSE_estimator

**Repository Path**: mtxing/Auditory_model_based_MMSE_estimator

## Basic Information

- **Project Name**: Auditory_model_based_MMSE_estimator
- **Description**: Auditory_model_based_MMSE_estimator
去噪算法/语音增强方法

基于听觉模型的贝叶斯估计器用于语音增强

该功能实现了基于参数贝叶斯估计器的音频降噪算法。参数估计器的参数是根据人类的听觉特征（例如耳蜗的动态压缩非线性和响度感知理论）选择的。它使用具有低复杂度和低跟踪延迟的基于MMSE的无偏噪声估计

此功能将嘈杂的音频文件作为输入并生成增强的音频文件。

用法：Auditory_MMSE_STSA_unbiased_ne（noisyFile，enhancedFile，alpha，beta）

跑步：
运行程序demo_process_file.m

苏曼苏梅岛（c）2015版权所有（c）
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 1
- **Created**: 2021-06-07
- **Last Updated**: 2021-06-07

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Auditory_model_based_MMSE_estimator
A denoising algorithm/ Speech Enhancement method


 Auditory-model based Bayesian estimator for Speech enhancement
 
 This function implements an audio denoising algorithm based o paramaterized 
 Bayesian estimator. The parameters The parameters of the estimator are chosen 
 based on the human auditory characteristics such as the cochlea’s dynamic
 compressive non-linearity and loudness perception theory. It uses an unbiased 
 MMSE-Based Noise Power Estimation with Low Complexity and Low Tracking Delay
   
 This function takes noisy audio file as input and generates an enhanced audio file.  

 Usage:  Auditory_MMSE_STSA_unbiased_ne(noisyFile,enhancedFile,alpha,beta)
 
 
 - To run:
 
 run the program demo_process_file.m

 Copyright (c) 2015 by Suman Samui