{"id":24774402,"url":"https://github.com/unb-libraries/sebastian","last_synced_at":"2025-03-23T22:45:34.862Z","repository":{"id":274381141,"uuid":"922723732","full_name":"unb-libraries/sebastian","owner":"unb-libraries","description":"Sebastian transcribes audio files.","archived":false,"fork":false,"pushed_at":"2025-03-17T14:42:39.000Z","size":342,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":5,"default_branch":"master","last_synced_at":"2025-03-17T15:51:25.360Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/unb-libraries.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":"2025-01-26T23:25:05.000Z","updated_at":"2025-03-17T14:42:43.000Z","dependencies_parsed_at":"2025-01-27T00:23:51.062Z","dependency_job_id":"2252f313-2464-45ac-b888-ef59794a9e8f","html_url":"https://github.com/unb-libraries/sebastian","commit_stats":null,"previous_names":["unb-libraries/sebastian"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/unb-libraries%2Fsebastian","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/unb-libraries%2Fsebastian/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/unb-libraries%2Fsebastian/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/unb-libraries%2Fsebastian/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/unb-libraries","download_url":"https://codeload.github.com/unb-libraries/sebastian/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":245181541,"owners_count":20573717,"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":[],"created_at":"2025-01-29T05:48:47.368Z","updated_at":"2025-03-23T22:45:34.848Z","avatar_url":"https://github.com/unb-libraries.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# sebastian\n\u003cp align=\"center\"\u003e\n\u003cimg src=\"assets/image.webp\" alt=\"drawing\" width=\"400\"/\u003e\n\u003c/p\u003e\n\n## Introduction\nSebastian transcribes audio files to text. It leverages the [whisperX](https://github.com/m-bain/whisperX) library, a speech-to-text library that extends Whisper ASR model [originally developed by OpenAI](https://github.com/openai/whisper).\n\nBoth time-alignment and diarization (unique speaker identification) are supported.\n\n## Documentation\nSome stub documentation is currently available [in the documentation folder](./documentation/README.md \"Project Documentation\").\n\n## Pre-requisites\n- Python 3.8+\n- [Poetry](https://python-poetry.org/docs/)\n- [Nvidia drivers](https://www.nvidia.com/Download/index.aspx)\n- [CUDA](https://developer.nvidia.com/cuda-downloads)\n\n## CLI Commands\n### `api:start`\n\nStart the API server:\n\n`poetry run api:start`\n\n## Client Request Format\n\nThe client should send a POST request with multipart form data containing an audio file. The form data can contain the following fields:\n\n### Standard Transcriptions\n\n- **file**: The audio file to be transcribed. This is a required field.\n- **model**: The whisper model to be used for transcription. Default is 'large-v3-turbo'.\n- **device**: The device to be used for computation. Default is 'cuda'.\n- **compute_type**: The precision of computation to be used. Default is 'float16'. int8 is also supported but not recommended.\n- **batch_size**: The batch size for processing. Default is '16'.\n- **language**: The language of the audio file. Default is 'en'.\n\n### Time-Alignment\nThe transcription can also be time-aligned. The form data can contain the following additional fields:\n- **align**: A boolean flag to enable or disable alignment. Default is 'false'.\n\n### Diarization\nThe transcription can also be diarized to identify unique speakers in the transcription (SPEAKER-1, SPEAKER-2, etc.). The form data can contain the following additional fields:\n- **diarize**: A boolean flag to enable or disable diarization. Default is 'false'.\n- **min_speakers**: The minimum number of speakers to identify. Default is '1'.\n- **max_speakers**: The maximum number of speakers to identify. Default is '4'.\n\n### Debugging\n- **debug**: A boolean flag to enable or disable debug mode. Default is 'false'.\n\nThe API will return a JSON response with the transcribed text data\n\n## Convenience Commands\nOther commands are available for convenience:\n\n### `transcribe`\nThe input file path should be an absolute path to an audio file you wish to summarize.\n\n`poetry run transcribe test.mp4`\n\n## License\n- In line with our 'open' ethos, UNB Libraries makes its applications and workflows freely available to everyone whenever possible.\n- As a result, the contents of this repository [unb-libraries/sebastian] are licensed under the [MIT License](http://opensource.org/licenses/mit-license.html). This license explicitly excludes:\n   - Any content that remains the exclusive property of its author(s).\n   - The UNB logo and any associated visual identity assets remain the exclusive property of the University of New Brunswick.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Funb-libraries%2Fsebastian","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Funb-libraries%2Fsebastian","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Funb-libraries%2Fsebastian/lists"}