{"id":13659884,"url":"https://github.com/richardassar/SampleRNN_torch","last_synced_at":"2025-04-24T15:31:09.160Z","repository":{"id":145354995,"uuid":"87479491","full_name":"richardassar/SampleRNN_torch","owner":"richardassar","description":"Torch implementation of SampleRNN: An Unconditional End-to-End Neural Audio Generation Model","archived":false,"fork":false,"pushed_at":"2017-05-21T22:22:43.000Z","size":56,"stargazers_count":156,"open_issues_count":5,"forks_count":24,"subscribers_count":8,"default_branch":"master","last_synced_at":"2024-11-10T13:37:09.169Z","etag":null,"topics":["audio","deep-learning","generative","neural-network","torch"],"latest_commit_sha":null,"homepage":"","language":"Lua","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/richardassar.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null}},"created_at":"2017-04-06T22:08:37.000Z","updated_at":"2024-01-04T16:12:59.000Z","dependencies_parsed_at":null,"dependency_job_id":"60167fbb-df59-474c-a8ce-cc6f218ffa4d","html_url":"https://github.com/richardassar/SampleRNN_torch","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/richardassar%2FSampleRNN_torch","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/richardassar%2FSampleRNN_torch/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/richardassar%2FSampleRNN_torch/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/richardassar%2FSampleRNN_torch/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/richardassar","download_url":"https://codeload.github.com/richardassar/SampleRNN_torch/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":250654193,"owners_count":21465833,"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","deep-learning","generative","neural-network","torch"],"created_at":"2024-08-02T05:01:13.272Z","updated_at":"2025-04-24T15:31:08.646Z","avatar_url":"https://github.com/richardassar.png","language":"Lua","readme":"# SampleRNN_torch\r\n\r\nA Torch implementation of [SampleRNN: An Unconditional End-to-End Neural Audio Generation Model](https://openreview.net/forum?id=SkxKPDv5xl).\r\n\r\n![A visual representation of the SampleRNN architecture](http://deepsound.io/images/samplernn.png)\r\n\r\n## Samples\r\n\r\nListen to a selection of generated output at the following links:\r\n\r\n- [piano](https://soundcloud.com/psylent-v/sets/samplernn_torch)\r\n- [mozart](https://soundcloud.com/psylent-v/sets/samplernn_torch-mozart)\r\n\r\nFeel free to submit links to any interesting output you generate or dataset creation scripts as a pull request.\r\n\r\n## Dependencies\r\n\r\nThe following packages are required to run SampleRNN_torch:\r\n\r\n- nn\r\n- cunn\r\n- cudnn\r\n- rnn\r\n- optim\r\n- audio\r\n- xlua\r\n- gnuplot\r\n\r\n**NOTE**: Update `nn` and `cudnn` even if they were already installed as fixes have been submitted which affect this project.\r\n\r\n## Datasets\r\n\r\nTo retrieve and prepare the *piano* dataset, as used in the reference implementation, run:\r\n\r\n```\r\ncd datasets/piano/\r\n./create_dataset.sh\r\n```\r\n\r\nOther dataset preparation scripts may be found under `datasets/`.\r\n\r\nCustom datasets may be created by using `scripts/generate_dataset.lua` to slice multiple audio files into segments for training, audio must be placed in `datasets/[dataset]/data/`.\r\n\r\n## Training\r\n\r\nTo start a training session run `th train.lua -dataset piano`. To view a description of all accepted arguments run `th train.lua -help`.\r\n\r\nTo view the progress of training run `th generate_plots`, the loss and gradient norm curve will be saved in `sessions/[session]/plots/`.\r\n\r\n## Sampling\r\n\r\nBy default samples are generated at the end of every training epoch but they can also be generated separately using `th train.lua -generate_samples` with the `session` parameter to specify the model.\r\n\r\nMultiple samples are generated in batch mode for efficiency, however generating a single audio sample is faster with `th fast_sample.lua`. See `-help` for a description of the arguments.\r\n\r\n## Models\r\n\r\nA pretrained model of the *piano* dataset is available [here](https://drive.google.com/uc?id=0B5pXFO5X-KJ9Mko3MUZuLUpEQVU\u0026export=download). Download and copy it into your `sessions/` directory and then extract it in place.\r\n\r\nMore models will be uploaded soon.\r\n\r\n## Theano version\r\n\r\nThis code is based on the reference implementation in Theano.\r\n\r\nhttps://github.com/soroushmehr/sampleRNN_ICLR2017\r\n","funding_links":[],"categories":["Lua","Model Zoo"],"sub_categories":["Recurrent Networks"],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Frichardassar%2FSampleRNN_torch","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Frichardassar%2FSampleRNN_torch","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Frichardassar%2FSampleRNN_torch/lists"}