{"id":23725976,"url":"https://github.com/voidful/speechmix","last_synced_at":"2025-09-04T02:31:19.618Z","repository":{"id":40367831,"uuid":"422627804","full_name":"voidful/SpeechMix","owner":"voidful","description":"Explore different way to mix speech model(wav2vec2, hubert) and nlp model(BART,T5,GPT) together","archived":false,"fork":false,"pushed_at":"2025-07-03T06:31:43.000Z","size":97,"stargazers_count":47,"open_issues_count":3,"forks_count":10,"subscribers_count":4,"default_branch":"main","last_synced_at":"2025-08-17T02:19:32.504Z","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":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/voidful.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,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2021-10-29T15:35:41.000Z","updated_at":"2025-07-03T06:31:47.000Z","dependencies_parsed_at":"2025-07-03T07:37:29.511Z","dependency_job_id":null,"html_url":"https://github.com/voidful/SpeechMix","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/voidful/SpeechMix","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/voidful%2FSpeechMix","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/voidful%2FSpeechMix/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/voidful%2FSpeechMix/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/voidful%2FSpeechMix/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/voidful","download_url":"https://codeload.github.com/voidful/SpeechMix/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/voidful%2FSpeechMix/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":273541897,"owners_count":25124056,"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":"2025-09-04T02:00:08.968Z","response_time":61,"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":"2024-12-31T00:18:03.189Z","updated_at":"2025-09-04T02:31:19.343Z","avatar_url":"https://github.com/voidful.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# SpeechMix\n\nExplore different way to mix speech model(wav2vec2, hubert) and nlp model(BART,T5,GPT) together.   \n\nImplementation of:   \n[Large-Scale Self- and Semi-Supervised Learning for Speech Translation](https://arxiv.org/abs/2104.06678) - ACL2021   \n[Multilingual Speech Translation with Efficient Finetuning of Pretrained Models](https://arxiv.org/abs/2010.12829) - ACL2021   \n[Lightweight Adapter Tuning for Multilingual Speech Translation](https://arxiv.org/abs/2106.01463) - Interspeech 2021   \n[Improving Speech Translation by Understanding and Learning from the Auxiliary Text Translation Task](https://arxiv.org/abs/2107.05782) - ACL2021   \n[A General Multi-Task Learning Framework to Leverage Text Data for Speech to Text Tasks](https://arxiv.org/abs/2010.11338) - ICASSP 2021   \n\n## Installation\n\n### pip install\n\n```bash\npip install speechmix\n```\n\n### Build from source\n\ngit clone and cd into this project.\n\n```shell\npip install -e .\n```\n\n## Name the project(!important)\nWANDB_PROJECT=amazing\n\n## base\n```shell\npython train.py --speech_model_config wav2vec2 \\\n--nlp_model_config facebook/bart-base \\\n--SpeechMixEED \\\n--dataset librispeech_asr \\\n--field clean \\\n--train_split train.100 \\\n--test_split validation \\\n--batch 3 \\\n--grad_accum 20 \\\n--epoch 30 \\\n--worker 15 \\\n--share_layer_ratio 0 \\\n--down_scale 2 \\\n--lr 4e-5 \\\n--warmup_steps 500 \\\n--wandb \\\n--notes base\n```\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvoidful%2Fspeechmix","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fvoidful%2Fspeechmix","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvoidful%2Fspeechmix/lists"}