Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/fgnt/ci_sdr

https://github.com/fgnt/ci_sdr

Science Score: 41.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ○
    codemeta.json file
  • ○
    .zenodo.json file
  • ✓
    DOI references
    Found 4 DOI reference(s) in README
  • ✓
    Academic publication links
    Links to: arxiv.org
  • ✓
    Committers with academic emails
    1 of 3 committers (33.3%) from academic institutions
  • ✓
    Institutional organization owner
    Organization fgnt has institutional domain (nt.uni-paderborn.de)
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (13.2%) to scientific vocabulary
Last synced: 6 months ago · JSON representation

Repository

Basic Info
Statistics
  • Stars: 52
  • Watchers: 5
  • Forks: 8
  • Open Issues: 0
  • Releases: 0
Created about 5 years ago · Last pushed 11 months ago
Metadata Files
Readme License

README.md

Convolutive Transfer Function Invariant SDR

ISCA DOI arXiv

Run python tests PyPI codecov.io PyPI - Downloads License: MIT

This repository contains an implementation for the Convolutive transfer function Invariant Signal-to-Distortion Ratio objective for PyTorch as described in the publication Convolutive Transfer Function Invariant SDR training criteria for Multi-Channel Reverberant Speech Separation (IEEE ICASSP, RIS UPB, arXiv).

Here, a small example, how you can use this CI-SDR objective in your own source code:

```python import torch import ci_sdr

reference: torch.tensor = ...

reference.shape: [speakers, samples]

estimation: torch.tensor = ...

estimation shape: [speakers, samples]

sdr = cisdr.pt.cisdr_loss(estimation, reference)

sdr shape: [speakers]

```

The idea of this objective function is based in the theory from E. Vincent, R. Gribonval and C. Févotte, Performance measurement in blind audio source separation, IEEE Trans. Audio, Speech and Language Processing, known as BSSEval. The original author provided MATLAB source code (link) and the package mir_eval (link) contains a python port. Some people refer to these implementations as BSSEval v3 (link).

The PyTorch code in this package is tested to yield the same SDR values as mir_eval with the default parameters.

NOTE: If you want to use BSSEval v3 SDR as metric, I recomment to use mir_eval.separation.bss_eval_sources and use as reference the clean/unreverberated source signals. The implementation in this repository has minor difference that makes it problematic to compare SDR values accorss different publications (e.g. here the permutation is calculated on the SDR, while mir_eval computes it based on the SIR.).

Installation

Install it directly with Pip, if you just want to use it:

bash pip install ci-sdr

or to get the recent version:

bash pip install git+https://github.com/fgnt/ci_sdr.git

If you want to install it with all dependencies (test and doctest dependencies), run:

bash pip install git+https://github.com/fgnt/ci_sdr.git#egg=ci_sdr[all]

When you want to change the code, clone this repository and install it as editable:

```bash git clone https://github.com/fgnt/cisdr.git cd cisdr pip install --editable .

pip install --editable .[all]

```

Citation

To cite this implementation, you can cite the following paper (IEEE ICASSP, RIS UPB, arXiv): @inproceedings{boeddeker2021convolutive, title={Convolutive transfer function invariant SDR training criteria for multi-channel reverberant speech separation}, author={Boeddeker, Christoph and Zhang, Wangyou and Nakatani, Tomohiro and Kinoshita, Keisuke and Ochiai, Tsubasa and Delcroix, Marc and Kamo, Naoyuki and Qian, Yanmin and Haeb-Umbach, Reinhold}, booktitle={IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, pages={8428--8432}, year={2021}, organization={IEEE}, doi={10.1109/ICASSP39728.2021.9414661}} }

Owner

  • Name: Department of Communications Engineering University of Paderborn
  • Login: fgnt
  • Kind: organization
  • Location: Paderborn, Germany

GitHub Events

Total
  • Issues event: 1
  • Issue comment event: 2
  • Push event: 10
  • Pull request event: 1
  • Pull request review event: 4
  • Pull request review comment event: 1
  • Create event: 1
Last Year
  • Issues event: 1
  • Issue comment event: 2
  • Push event: 10
  • Pull request event: 1
  • Pull request review event: 4
  • Pull request review comment event: 1
  • Create event: 1

Committers

Last synced: almost 3 years ago

All Time
  • Total Commits: 68
  • Total Committers: 3
  • Avg Commits per committer: 22.667
  • Development Distribution Score (DDS): 0.029
Past Year
  • Commits: 5
  • Committers: 1
  • Avg Commits per committer: 5.0
  • Development Distribution Score (DDS): 0.0
Top Committers
Name Email Commits
Christoph Boeddeker c****j@m****e 66
jensheit h****r@n****e 1
neumann t****n@m****e 1
Committer Domains (Top 20 + Academic)
mail.upb.de: 1 nt.upb.de: 1 mail.uni-paderborn.de: 1

Issues and Pull Requests

Last synced: 8 months ago

All Time
  • Total issues: 3
  • Total pull requests: 4
  • Average time to close issues: 9 months
  • Average time to close pull requests: 1 day
  • Total issue authors: 2
  • Total pull request authors: 1
  • Average comments per issue: 2.67
  • Average comments per pull request: 0.75
  • Merged pull requests: 4
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 0
  • Pull requests: 1
  • Average time to close issues: N/A
  • Average time to close pull requests: about 24 hours
  • Issue authors: 0
  • Pull request authors: 1
  • Average comments per issue: 0
  • Average comments per pull request: 1.0
  • Merged pull requests: 1
  • Bot issues: 0
  • Bot pull requests: 0
View more stats
Top Authors
Issue Authors
  • kamo-naoyuki (2)
  • yunzqq (1)
Pull Request Authors
  • boeddeker (5)
Top Labels
Issue Labels
Pull Request Labels

Packages

  • Total packages: 3
  • Total downloads:
    • pypi 14,099 last-month
  • Total docker downloads: 706
  • Total dependent packages: 3
    (may contain duplicates)
  • Total dependent repositories: 8
    (may contain duplicates)
  • Total versions: 5
  • Total maintainers: 3
pypi.org: ci-sdr

A sample Python project

  • Versions: 2
  • Dependent Packages: 2
  • Dependent Repositories: 8
  • Downloads: 14,099 Last month
  • Docker Downloads: 706
Rankings
Downloads: 2.3%
Docker downloads count: 2.5%
Dependent packages count: 3.2%
Dependent repos count: 5.2%
Average: 5.8%
Stargazers count: 9.5%
Forks count: 11.9%
Maintainers (2)
cbj thequilo
Last synced: 6 months ago
spack.io: py-ci-sdr

This repository contains an implementation for the Convolutive transfer function Invariant Signal-to-Distortion Ratio objective for PyTorch as described in the publication Convolutive Transfer Function Invariant SDR training criteria for Multi-Channel Reverberant Speech Separation

  • Versions: 1
  • Dependent Packages: 0
  • Dependent Repositories: 0
Rankings
Dependent repos count: 0.0%
Stargazers count: 21.3%
Forks count: 25.1%
Average: 25.9%
Dependent packages count: 57.3%
Maintainers (1)
adamjstewart
Last synced: 6 months ago
conda-forge.org: ci-sdr

An implementation for the Convolutive transfer function Invariant Signal-to-Distortion Ratio objective for PyTorch as described in the publication Convolutive Transfer Function Invariant SDR training criteria for Multi-Channel Reverberant Speech Separation.

  • Versions: 2
  • Dependent Packages: 1
  • Dependent Repositories: 0
Rankings
Dependent packages count: 28.8%
Dependent repos count: 34.0%
Average: 36.0%
Stargazers count: 37.7%
Forks count: 43.4%
Last synced: 7 months ago

Dependencies

setup.py pypi
  • einops *
  • numpy *
  • scipy *
  • torch *