VoiceSplit - GitLite

KO | EN

gitlite — search

#python #hacktoberfest #react #llm #machine-learning #java #windows #awesome #cli #kubernetes #awesome-list #vue #ios #mcp

VoiceSplit

(★ 271)

VoiceSplit: Targeted Voice Separation by Speaker-Conditioned Spectrogram

파일 탐색기

최종 버전 다운로드 (.zip)

- - dev.csv
  - README.md
  - test_config.csv
  - test_demo.csv
  - train.csv
  - train_small.csv
- - dev.csv
- - model.py
- - model.py
- embedder.pt
- GE2E-CorentinJ-ExtractSpeakerEmbedding.ipynb
- GE2E-Seungwonpark-ExtractSpeakerEmbedding-adaptado-para-openvoicefilter.ipynb
- GE2E-Seungwonpark-ExtractSpeakerEmbedding-adaptado-para-openvoicefilter.py
- GE2E-Seungwonpark-ExtractSpeakerEmbedding.ipynb
- nohup.out
- Speech2Phone-ExtractSpeakerEmbeddings.ipynb
- generate_VCTK_dev_csv.py
- normalise-resample.sh
- resample-audio-files.sh
- test_extract_audio.py
- test_extract_audio_torch.py
- .goutputstream-LQ55M0
- .goutputstream-YRB4M0
- audio.py
- audio_processor.py
- dataset.py
- generic_utils.py
- stft.py
- tensorboard.py
.gitignore
config.json
convert.py
Final Report.pdf
generator_paper.py
LICENSE
log-out.txt
Partial Report.pdf
preprocess_by_csv.py
preprocess_by_csv_without_voice_overlay.py
README.md
requirements.txt
run_train.sh
run_train.sh.save
test.py
test_all_checkpoints.py
test_fast_all_checkpoints.py
train.py
VoiceSplit presentation.pdf

# 설치 가이드

1. 코드 내려받기

git clone https://github.com/Edresson/VoiceSplit

깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.

cd VoiceSplit

방금 내려받은 프로젝트 폴더 안으로 이동합니다.

2. Python

쉬움 추천

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Python 3 설치 시 'Add Python to PATH' 옵션을 꼭 체크하세요 (Windows).

pip install -r requirements.txt

requirements.txt 등에 명시된 파이썬 라이브러리를 설치합니다.

jupyter notebook

브라우저에서 노트북(.ipynb) 파일들을 열람하고 실행할 수 있는 Jupyter 화면을 켭니다.

✅ 에러 메시지 없이 실행되고 터미널에 안내 문구가 출력되면 정상입니다.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# VoiceSplit

Pytorch unofficial implementation of [VoiceFilter: Targeted Voice Separation by Speaker-Conditioned Spectrogram Masking](https://arxiv.org/abs/1810.04826)

Final project for **SCC5830- Image Processing** @ ICMC/USP.

## Dataset
For the task we intend to use the LibriSpeech dataset initially. However, to use it in this task, we need to generate audios with overlappings voices.

## Improvements
    We use Si-SNR with PIT instead of Power Law compressed loss, because it allows us to achieve a better result ( comparison available in: https://github.com/Edresson/VoiceSplit).
    We used the MISH activation function instead of ReLU and this has improved the result

# Report
You can see a report of what was done in this repository [here](https://github.com/Edresson/VoiceSplit/blob/master/Final%20Report.pdf)

## Demos
Colab notebooks Demos:

Exp 1: [link](https://colab.research.google.com/drive/1GljrJOo_uMRfSUDIIht1Eo1HqaAkhY36?usp=sharing)

Exp 2: [link](https://colab.research.google.com/drive/19Rh4YaZtcI2gSAvI9q40YUXubxOXLkUG?usp=sharing)

Exp 3: [link](https://drive.google.com/file/d/1b0gCNazK6exulR765PAI3xkMMXzBR-Fp/view?usp=sharing)

Exp 4: [link](https://colab.research.google.com/drive/1AJ0VS8_Vv3Ayph20WFyi6iWASEJpQE3N?usp=sharing)

Exp 5 (best): [link](https://colab.research.google.com/drive/1FYlaQX4XzN4W_6JB4z9R5SFrBF8a9RqC?usp=sharing)

Site demo for the experiment with best results (Exp 5): https://edresson.github.io/VoiceSplit/
## ToDos:
Create documentation for the repository and
remove unused code

## Future Works

* Train VoiceSplit model with GE2E3k and Mean Squared Error loss function

## Acknowledgment:
In this repository it contains codes of other collaborators, the due credits were given in the used functions:

Preprocessing: Eren Gölge @erogol

VoiceFilter Model: Seungwon Park @seungwonpark