{ "id": 307959, "title": "state-of-the-art & most popular sound research", "url": "/competitions/birdclef-2022/discussion/307959", "author_name": "", "post_date": "2022-02-16T12:12:52.477313900Z", "votes": 30, "comment_count": 3, "views": 0, "content": "", "messages": [ { "id": "1693055", "postDate": "02/16/2022 12:12:52", "content": "", "rawMarkdown": "### State-of-the-art:\n**Lung Sound Classification Using Co-tuning and Stochastic Normalization**\nPaper: https://arxiv.org/pdf/2108.01991v1.pdf\nCode: https://github.com/makcedward/nlpaug\n**Efficient Training of Audio Transformers with Patchout** (PaSST)\nPaper: https://arxiv.org/pdf/2110.05069v2.pdf\nCode:https://github.com/kkoutini/passt\n**AST: Audio Spectrogram Transformer** (AST)\nPaper: https://arxiv.org/pdf/2110.05069v2.pdf\nCode: https://github.com/kkoutini/passt\n**Environmental Sound Classification on the Edge: A Pipeline for Deep Acoustic Networks on Extremely Resource-Constrained Devices** (ACEDet)\nPaper: https://arxiv.org/pdf/2103.03483v3.pdf\nCode: https://github.com/mohaimenz/acdnet\n\n\n\n\n\n\n\n- ### Most popular:\n**Deep Convolutional Neural Networks and Data Augmentation for Environmental Sound Classification**\nMain Idea: CNN for frequency feature extraction\nPaper: https://arxiv.org/pdf/1608.04363v2.pdf\nCode: https://github.com/justinsalamon/UrbanSound8K-JAMS\nPopular Implement: https://github.com/makcedward/nlpaug\n\n- ### Relative:\n**Recognizing Birds from Sound - The 2018 BirdCLEF Baseline System**\nPaper: https://arxiv.org/pdf/1804.07177v1.pdf\nCode: https://github.com/kahst/BirdCLEF-Baseline\n\n\n\n\n\n\n- ### Popular Datasets:\nAudioSet ESC-50 UrbanSound8K FSD50K DiCOVA FSDnoisy18k YouTube-100M RAVDESS", "votes": null }, { "id": "1695247", "postDate": "02/18/2022 03:20:17", "content": "

good job man!

", "rawMarkdown": "good job man!", "votes": null }, { "id": "1695271", "postDate": "02/18/2022 04:01:03", "content": "

You are welcome : )

", "rawMarkdown": "You are welcome : )", "votes": null }, { "id": "1796058", "postDate": "05/20/2022 12:34:21", "content": "

thank you！

", "rawMarkdown": "thank you！", "votes": null } ], "comments": [ { "id": 1695247, "author_name": "leolan12", "author_url": "", "post_date": "02/18/2022 03:20:17", "content": "

good job man!

", "votes": null, "replies": [ { "id": 1695271, "author_name": "dwchen", "author_url": "", "post_date": "02/18/2022 04:01:03", "content": "

You are welcome : )

", "votes": null, "replies": [] } ] }, { "id": 1796058, "author_name": "jintarokawai", "author_url": "", "post_date": "05/20/2022 12:34:21", "content": "

thank you！

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1693055": "### State-of-the-art:\n**Lung Sound Classification Using Co-tuning and Stochastic Normalization**\nPaper: https://arxiv.org/pdf/2108.01991v1.pdf\nCode: https://github.com/makcedward/nlpaug\n**Efficient Training of Audio Transformers with Patchout** (PaSST)\nPaper: https://arxiv.org/pdf/2110.05069v2.pdf\nCode:https://github.com/kkoutini/passt\n**AST: Audio Spectrogram Transformer** (AST)\nPaper: https://arxiv.org/pdf/2110.05069v2.pdf\nCode: https://github.com/kkoutini/passt\n**Environmental Sound Classification on the Edge: A Pipeline for Deep Acoustic Networks on Extremely Resource-Constrained Devices** (ACEDet)\nPaper: https://arxiv.org/pdf/2103.03483v3.pdf\nCode: https://github.com/mohaimenz/acdnet\n\n\n\n\n\n\n\n- ### Most popular:\n**Deep Convolutional Neural Networks and Data Augmentation for Environmental Sound Classification**\nMain Idea: CNN for frequency feature extraction\nPaper: https://arxiv.org/pdf/1608.04363v2.pdf\nCode: https://github.com/justinsalamon/UrbanSound8K-JAMS\nPopular Implement: https://github.com/makcedward/nlpaug\n\n- ### Relative:\n**Recognizing Birds from Sound - The 2018 BirdCLEF Baseline System**\nPaper: https://arxiv.org/pdf/1804.07177v1.pdf\nCode: https://github.com/kahst/BirdCLEF-Baseline\n\n\n\n\n\n\n- ### Popular Datasets:\nAudioSet ESC-50 UrbanSound8K FSD50K DiCOVA FSDnoisy18k YouTube-100M RAVDESS", "1695247": "good job man!", "1695271": "You are welcome : )", "1796058": "thank you！" }, "source": "meta" }