hyli666 / dnn-speechenhancement

Welcome to GitHub Pages

You can use the editor on GitHub to maintain and preview the content for your website in Markdown files.

Whenever you commit to this repository, GitHub Pages will run Jekyll to rebuild the pages in your site, from the content in your Markdown files.

Markdown

Markdown is a lightweight and easy-to-use syntax for styling your writing. It includes conventions for

Syntax highlighted code block

# Header 1
## Header 2
### Header 3

- Bulleted
- List

1. Numbered
2. List

**Bold** and _Italic_ and `Code` text

[Link](url) and ![Image](src)

For more details see GitHub Flavored Markdown.

Jekyll Themes

Your Pages site will use the layout and styles from the Jekyll theme you have selected in your repository settings. The name of this theme is saved in the Jekyll _config.yml configuration file.

Support or Contact

Having trouble with Pages? Check out our documentation or contact support and we’ll help you sort it out.

dnn-speechenhancement's Issues

How to get the mvn_store.mat? And what's its meaning?

Dr Li:
Hi, I'm studying your code these days, and I still have some problems. Please give me some advice!

1. How to get the mvn_store.mat? And what's its meaning?
2. How to transfer *.wav to a matrix?
3. In DNN_sgd_lps.py, 221st lines, where's "_nat" come from? and "_nat" is an error when I tried to run the code.
   Hope for your reply.

Data files format

Hi,
Good job for converting the project to TF.
Can you upload your scrips for generating the data matlab files, or give a short explanation on what is in each file.
I tried to figgure jt out, and this is what i think hoing in there, please correct me if somthing is wrong:
the train data+label is pre normalized.
The label is 2572 = clean data + original noise
The data is 2578 = 3 past + present+3 future+ noise estimation
The validation part is loaded in different format, as raw spectrogram, and in the code normalize+add noise estimation = 8257
And validation label is original clean data + original noise = 2572

Thanks.

关于带噪数据生成的问题？

你好，hyli666.
最近看了一些论文，开始着手代码实践部分，您的genRawData_func.m代码中带噪数据总是生成不了，我用的数据集是NOISEX-92中的噪声数据，干净语音使用的是NOISE中的clean数据，不知道是不是我传入的数据的问题，这个genrawdata代码在调用addnoise时总会提示语音信号的长度大于噪声长度，我尝试修改了采样率以及数据长度(以补零的方式增加了噪声长度)，运行之后又提示各种问题，索引维度什么的。如果是我传入数据的问题，希望您可以指出对输入数据的一些要求比如采样率或者信号长度。另外您可不可以将您的训练数据上传一部分。

期待您的回复————刚接触语音增强的小白

语音增强小白

mvn_store.mat如何生成？

1.mvn_store.mat是您直接上传的数据，请问如果我更改训练数据噪声后应该如何生成mvn_store.mat？mvn_store.mat里面的global_mean和global_std分别指的是什么？
2.DNN_sgd_lps.py文件中，权值w和偏置b如何更新的？在代码哪里体现出来的？
3.徐勇论文用的是dbn模型，您的DNN_sgd_lps.py文件代码里面体现第一层rbm的地方在哪？