phasen - GitLite

KO | EN

gitlite — search

#hacktoberfest #llm #golang #go #android #machine-learning #rust #deep-learning #linux #nodejs #windows #cli #ios #openai

phasen

(★ 235)

A unofficial Pytorch implementation of Microsoft's PHASEN

File Explorer

Download Latest Version (.zip)

- - cv_wsj0_-5~20_mix.lst
  - test_wsj0_mix_-5.lst
  - test_wsj0_mix_0.lst
  - test_wsj0_mix_10.lst
  - test_wsj0_mix_15.lst
  - test_wsj0_mix_20.lst
  - test_wsj0_mix_5.lst
  - tr_wsj0_-5~20_mix.lst
- cv_wsj0_-5~20.lst
- t
- test_wsj0_-5.lst
- test_wsj0_0.lst
- test_wsj0_10.lst
- test_wsj0_15.lst
- test_wsj0_20.lst
- test_wsj0_5.lst
- tr_wsj0_-5~20.lst
- noisy.png
- phasen-net.png
- phasen.png
- conv_stft.py
- old.py
- phasen.py
- show.py
- run_phasen.py
- __init__.py
- add_duration.py
- add_noise.py
- eval_objective.py
- get_results.py
- misc.py
- prepare_test.sh
- prepare_train.sh
- si_sdr_numpy.py
- t
- time_dataset.py
.gitignore
README.md
requirements.txt
run_phasen.sh

# Installation Guide

1. Get the code

git clone https://github.com/huyanxin/phasen

Downloads the entire project code from GitHub to your computer.

cd phasen

Moves into the project folder you just downloaded.

2. Python

Easy Recommended

Prerequisites

Git Needed to download the project code from GitHub.
Python 3 On Windows, be sure to check 'Add Python to PATH' during installation.

pip install -r requirements.txt

Installs the Python libraries listed in requirements.txt (or similar).

✅ If it runs without errors and prints output in the terminal, it worked.

Pulled directly from this repo's README.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# PHASEN

---
Unofficial PyTorch implementation of MSRA's:
    [PHASEN: A Phase-and-Harmonics-Aware Speech Enhancement Network](https://arxiv.org/abs/1911.04697).
![](https://raw.githubusercontent.com/huyanxin/phasen/master/figs/phasen-net.png)

---

## My resutls on real-world test
![Noisy](https://raw.githubusercontent.com/huyanxin/phasen/master/figs/noisy.png)
![enh](https://raw.githubusercontent.com/huyanxin/phasen/master/figs/phasen.png)

Maybe there is something different with the paper, but it worked not bad.

---

## how to use it?
1. install dependency:
```bash
pip install -r requirements.txt

```
2. download datasets

if you don't have WSJ0, you can follow this use aishell-1 by following this [se-cldnn-torch](https://github.com/huyanxin/se-cldnn-torch)

### Attetion
There is something different from [se-cldnn-torch](https://github.com/huyanxin/se-cldnn-torch):
the two list for train (tr.lst, cv.lst ...) need duration information, but [se-cldnn-torch](https://github.com/huyanxin/se-cldnn-torch) dose not need it (because the two dataset.py are different).

So, in this repo, train and cross-validation list nead to be like this
```
/path/noisy1.wav /path/ref1.wav 3.0233
/path/noisy2.wav /path/ref2.wav 2.3213
/path/noisy2.wav /path/ref2.wav 8.8127
...
```

To add duration information, you can use `tools/add_duration.py` like:

```
python tools/add_duration.py data/tr_wsj0.lst
```
As for inference stage (decode stage, eval stage), the list only need the path of noisy path:
```
/path/noisy1.wav
/path/noisy2.wav
/path/noisy2.wav
...
```

3. run.
before you run it, please set the correct params in `./run_phasen.sh`
```bash
bash run_phasen.sh
```

## Reference:
funcwj's [voice-filter](https://github.com/funcwj/voice-filter)

wangkenpu's [Conv-Tasnet](https://github.com/wangkenpu/Conv-TasNet-PyTorch)

pseeth's [torch-stft](https://github.com/pseeth/torch-stft)