{"id":15637919,"url":"https://github.com/dansuh17/segan-pytorch","last_synced_at":"2025-04-16T05:56:07.139Z","repository":{"id":61405413,"uuid":"113036554","full_name":"dansuh17/segan-pytorch","owner":"dansuh17","description":"SEGAN pytorch implementation https://arxiv.org/abs/1703.09452","archived":false,"fork":false,"pushed_at":"2019-03-11T06:27:09.000Z","size":84,"stargazers_count":108,"open_issues_count":8,"forks_count":32,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-03-29T05:05:01.394Z","etag":null,"topics":["audio","data-preprocessing","mir","pytorch","segan","segan-pytorch","source-separation","speech-enhancement"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/dansuh17.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}},"created_at":"2017-12-04T11:55:28.000Z","updated_at":"2025-02-03T08:02:49.000Z","dependencies_parsed_at":"2022-10-17T00:29:34.582Z","dependency_job_id":null,"html_url":"https://github.com/dansuh17/segan-pytorch","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dansuh17%2Fsegan-pytorch","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dansuh17%2Fsegan-pytorch/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dansuh17%2Fsegan-pytorch/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dansuh17%2Fsegan-pytorch/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/dansuh17","download_url":"https://codeload.github.com/dansuh17/segan-pytorch/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":249205592,"owners_count":21229954,"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","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":["audio","data-preprocessing","mir","pytorch","segan","segan-pytorch","source-separation","speech-enhancement"],"created_at":"2024-10-03T11:15:12.540Z","updated_at":"2025-04-16T05:56:07.122Z","avatar_url":"https://github.com/dansuh17.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Pytorch Implementation of SEGAN (Speech Enhancement GAN)\nImplementation of [SEGAN](https://arxiv.org/abs/1703.09452) by Pascual et al. in 2017, using pytorch.\nOriginal Tensorflow version can be found [here](https://github.com/santi-pdp/segan).\n\n## Prerequisites\n\n- python v3.5.2 or higher\n- pytorch v0.4.0\n- CUDA preferred\n- noisy speech dataset downloaded from [here](https://datashare.is.ed.ac.uk/handle/10283/2791)\n- libraries specified in `requirements.txt`\n\n### Installing Required Libraries\n\n`pip install -r requirements.txt`\n\n## Data Preprocessing\n\nUse `data_preprocess.py` file to preprocess downloaded data. \nAdjust the file paths at the beginning of the file to properly locate the data files, output folder, etc.\nUncomment functions in `__main__` to perform desired preprocessing stage.\n\nData preprocessing consists of three main stages:\n1. Downsampling - downsample original audio files (48k) to sampling rate of 16000.\n2. Serialization - Splitting the audio files into 2^14-sample (about 1 second) snippets.\n3. Verification - whether it contains proper number of samples.\n\nNote that the second stage takes a fairly long time - more than an hour.\n\n## Training\n\n```\npython model.py\n```\n\nAgain, fix and adjust datapaths in `model.py` according to your needs.\nEspecially, provide accurate path to where serialized data are stored.\n\n## Using Tensorboard\n\nIn order to use tensorboard, you need to first install tensorboard:\n\n```\npip install tensorboard\n```\n\nThen run tensorboard by specifing the log directory.\n\n```\ntensorboard --logdir=segan_data_out/tblogs\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdansuh17%2Fsegan-pytorch","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fdansuh17%2Fsegan-pytorch","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdansuh17%2Fsegan-pytorch/lists"}