{"id":19483634,"url":"https://github.com/diffapf/la-2a","last_synced_at":"2025-09-03T22:41:00.288Z","repository":{"id":232763496,"uuid":"774533820","full_name":"DiffAPF/LA-2A","owner":"DiffAPF","description":"Feed-forward compressor experiments source code for \"Differentiable All-pole Filters for Time-varying Audio Systems\".","archived":false,"fork":false,"pushed_at":"2024-06-10T16:14:09.000Z","size":47,"stargazers_count":21,"open_issues_count":0,"forks_count":1,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-08-29T16:08:16.997Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"https://diffapf.github.io/web/","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mpl-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/DiffAPF.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-03-19T18:02:05.000Z","updated_at":"2025-05-26T09:46:59.000Z","dependencies_parsed_at":"2024-04-17T10:37:38.313Z","dependency_job_id":"4bf609be-1970-492d-950a-cdb0ac6f8b1d","html_url":"https://github.com/DiffAPF/LA-2A","commit_stats":null,"previous_names":["diffapf/la-2a"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/DiffAPF/LA-2A","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DiffAPF%2FLA-2A","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DiffAPF%2FLA-2A/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DiffAPF%2FLA-2A/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DiffAPF%2FLA-2A/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/DiffAPF","download_url":"https://codeload.github.com/DiffAPF/LA-2A/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DiffAPF%2FLA-2A/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":273523634,"owners_count":25120864,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-09-03T02:00:09.631Z","response_time":76,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-10T20:16:10.347Z","updated_at":"2025-09-03T22:41:00.264Z","avatar_url":"https://github.com/DiffAPF.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cdiv align=\"center\"\u003e\n\u003ch1\u003eDifferentiable All-pole Filters for Time-varying Audio Systems\u003c/h1\u003e\n\n\u003cp\u003e\n    \u003ca href=\"https://yoyololicon.github.io/\" target=”_blank”\u003eChin-Yun Yu\u003c/a\u003e,\n    \u003ca href=\"https://christhetr.ee/\" target=”_blank”\u003eChristopher Mitcheltree\u003c/a\u003e,\n    \u003ca href=\"https://www.linkedin.com/in/alistair-carson-a6178919a/\" target=”_blank”\u003eAlistair Carson\u003c/a\u003e,\n    \u003ca href=\"https://www.acoustics.ed.ac.uk/group-members/dr-stefan-bilbao/\" target=”_blank”\u003eStefan Bilbao\u003c/a\u003e,\n    \u003ca href=\"https://www.eecs.qmul.ac.uk/~josh/\" target=”_blank”\u003eJoshua D. Reiss\u003c/a\u003e, and\n    \u003ca href=\"https://www.eecs.qmul.ac.uk/~gyorgyf/about.html\" target=”_blank”\u003eGyörgy Fazekas\u003c/a\u003e\n\u003c/p\u003e\n\n[![arXiv](https://img.shields.io/badge/arXiv-2404.07970-b31b1b.svg)](https://arxiv.org/abs/2404.07970)\n[![Listening Samples](https://img.shields.io/badge/%F0%9F%94%8A%F0%9F%8E%B6-Listening_Samples-blue)](https://diffapf.github.io/web/)\n[![Plugins](https://img.shields.io/badge/neutone-Plugins-blue)](https://diffapf.github.io/web/index.html#plugins)\n[![License](https://img.shields.io/badge/License-MPL%202.0-orange)](https://www.mozilla.org/en-US/MPL/2.0/FAQ/)\n\n\u003ch2\u003eFeed-forward Compressor (\u003cem\u003eLA-2A\u003c/em\u003e) Experiments\u003c/h2\u003e\n\u003c/div\u003e\n\n## Getting started\n\nFirst, please install the required packages, including our differentiable compressor [torchcomp](https://github.com/yoyololicon/torchcomp), by running:\n\n```bash\npip install -r requirements.txt\n```\n\n## Training\n\nFirstly, you need to download the SignalTrain dataset from [here](https://zenodo.org/records/3824876).\nThe training configurations are listed under `cfg/`.\nEach configurations listed under `cfg/data` corresponds to a dataset.\nPlease modify the `input` and `target` path of `cfg/data/la2a*.yaml` to the files of the dataset you downloaded.\n\nTo train the proposed differentiable feed-forward compressor, run:\n\n```bash\npython train_comp.py data=la2a_50\n```\nThe training logs will be uploaded to your wandb account under the project `dafx24`.\nIn this example, the model is trained with peak reduction of 50.\nChange the `data` argument to `la2a_75` or `la2a_25` to train the model with peak reduction of 75 or 25, respectively.\n\nTo train the frequency-sampling compressor (similar to [DASP](https://github.com/csteinmetz1/dasp-pytorch)), run:\n\n```bash\npython train_comp.py data=la2a_50 compressor.simple=true compressor.freq_sampling=true\n```\n\nA `ckpt.yaml` will be created under the logging folder (under `outputs/` by default) after training, which contains the parameters of the lowest training loss model.\nWe also provide our trained parameters under the folder `learned_params/`, with filenames as `[method]_[peak_reduction].yaml`.\n\n## Evaluation\n\nYou can use your checkpoints `ckpt.yaml` or our provided learned parameters to evaluate the compressor.\nGiven a wave file, you can compress it using the following command:\n\n```bash\npython test_comp.py ckpt.yaml input.wav output.wav\n```\n\n\n## Additional notes\n- `cfg/data/ff_*.yaml` are configurations for the feed-forward compressor experiments (FF-A/B/C in the paper). Please use `digital_compressor.py` to get the targets if you want to reproduce the experiments.\n\n## Links\n\n- [torchcomp](https://github.com/yoyololicon/torchcomp): Differentiable compressor implementation.\n- [training logs](https://wandb.ai/iamycy/torchcomp-la2a/): All training logs of the compressor experiments in the paper.\n\n## Citation\n\n```bibtex\n@inproceedings{ycy2024diffapf,\n  title={Differentiable All-pole Filters for Time-varying Audio Systems},\n  author={Chin-Yun Yu and Christopher Mitcheltree and Alistair Carson and Stefan Bilbao and Joshua D. Reiss and György Fazekas},\n  booktitle={International Conference on Digital Audio Effects (DAFx)},\n  year={2024}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdiffapf%2Fla-2a","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fdiffapf%2Fla-2a","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdiffapf%2Fla-2a/lists"}