{ "id": 197737, "title": "Previous Audio Competitions", "url": "/competitions/rfcx-species-audio-detection/discussion/197737", "author_name": "", "post_date": "2020-11-17T21:18:29.560396300Z", "votes": 88, "comment_count": 6, "views": 0, "content": "\n

Outside kaggle

\n", "messages": [ { "id": "1082388", "postDate": "11/17/2020 21:18:29", "content": "\n

Outside kaggle

\n", "rawMarkdown": "* **2020** https://www.kaggle.com/c/birdsong-recognition\n* **2019** https://www.kaggle.com/c/freesound-audio-tagging-2019\n* https://www.kaggle.com/c/the-icml-2013-bird-challenge\n* https://www.kaggle.com/c/mlsp-2013-birds\n* https://www.kaggle.com/c/whale-detection-challenge\n\n## Outside kaggle\n* 2020 https://www.aicrowd.com/challenges/lifeclef-2020-bird-monophone\n* 2019 https://www.crowdai.org/challenges/lifeclef-2019-bird-detection\n* 2018 http://dcase.community/challenge2018/task-bird-audio-detection", "votes": null }, { "id": "1082390", "postDate": "11/17/2020 21:23:31", "content": "

The recently finished Cornell Birdcall Identification top solutions collected by @addisonhoward

\n\n

Freesound Audio Tagging 2019 solutions collected by @jesucristo

\n", "rawMarkdown": "The recently finished **Cornell Birdcall Identification** top solutions collected by @addisonhoward \n- [1st Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183208)\n- [2nd Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183269)\n- [3rd Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183199)\n- [4th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183339)\n- [5th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183300)\n- [6th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183204)\n- [8th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183223)\n- [9th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183315)\n- [10th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183407)\n- [13th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183436)\n- [18th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183219)\n- [21st Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183209)\n- [36th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183222)\n- [39th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183258)\n- [43rd Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183240)\n\n**Freesound Audio Tagging** 2019 solutions collected by @jesucristo \n- https://www.kaggle.com/c/birdsong-recognition/discussion/158943", "votes": null }, { "id": "1082455", "postDate": "11/17/2020 22:54:41", "content": "

Also good to see DCASE challenges:
\nhttp://dcase.community/challenge2020/
\nit holds multiple challenges every year and participants submit technical reports which are quite helpful to know state-of-the-art techniques.

", "rawMarkdown": "Also good to see DCASE challenges:\nhttp://dcase.community/challenge2020/\nit holds multiple challenges every year and participants submit technical reports which are quite helpful to know state-of-the-art techniques.", "votes": null }, { "id": "1165769", "postDate": "01/23/2021 08:22:25", "content": "

Thanks for this list of links. Do you have an idea of most used features and models?

", "rawMarkdown": "Thanks for this list of links. Do you have an idea of most used features and models?", "votes": null }, { "id": "1166911", "postDate": "01/23/2021 22:32:03", "content": "

I did not checked the solutions in detail just had a glimpse at the descriptions.
\nWhat I learned from the Cornell comp:

\n\n

Both the Cornell and RFCX are multi class multilabel audio classification challenges for animal sounds. However the setup (soundscapes vs xeno canto, weird evaluation metrics, different train/test labels) is completely different. What worked there might not work here…

", "rawMarkdown": "I did not checked the solutions in detail just had a glimpse at the descriptions. \nWhat I learned from the Cornell comp:\n* MelSpectrogram with torchlibrosa\n* SED\n* PANNs\n* Synthetic noise augmentation\n* External Noise augmentation \n\nBoth the Cornell and RFCX are multi class multilabel audio classification challenges for animal sounds. However the setup (soundscapes vs xeno canto, weird evaluation metrics, different train/test labels) is completely different. What worked there might not work here...", "votes": null }, { "id": "1167289", "postDate": "01/24/2021 06:40:17", "content": "

Thanks a lot for the list. Indeed, some of the things have changed, I will keep that in mind. πŸ‘Œ

\n

I was planning to use librosa but now that you mention torchlibrosa, might be more suitable. πŸ‘

", "rawMarkdown": "Thanks a lot for the list. Indeed, some of the things have changed, I will keep that in mind. πŸ‘Œ \n\nI was planning to use librosa but now that you mention torchlibrosa, might be more suitable. πŸ‘", "votes": null }, { "id": "1175993", "postDate": "01/29/2021 12:49:26", "content": "

Thanks for this list of links. Do you have an idea of most used features and models?

", "rawMarkdown": "Thanks for this list of links. Do you have an idea of most used features and models?", "votes": null } ], "comments": [ { "id": 1082390, "author_name": "gaborfodor", "author_url": "", "post_date": "11/17/2020 21:23:31", "content": "

The recently finished Cornell Birdcall Identification top solutions collected by @addisonhoward

\n\n

Freesound Audio Tagging 2019 solutions collected by @jesucristo

\n", "votes": null, "replies": [] }, { "id": 1082455, "author_name": "hidehisaarai1213", "author_url": "", "post_date": "11/17/2020 22:54:41", "content": "

Also good to see DCASE challenges:
\nhttp://dcase.community/challenge2020/
\nit holds multiple challenges every year and participants submit technical reports which are quite helpful to know state-of-the-art techniques.

", "votes": null, "replies": [] }, { "id": 1165769, "author_name": "yassinealouini", "author_url": "", "post_date": "01/23/2021 08:22:25", "content": "

Thanks for this list of links. Do you have an idea of most used features and models?

", "votes": null, "replies": [ { "id": 1166911, "author_name": "gaborfodor", "author_url": "", "post_date": "01/23/2021 22:32:03", "content": "

I did not checked the solutions in detail just had a glimpse at the descriptions.
\nWhat I learned from the Cornell comp:

\n\n

Both the Cornell and RFCX are multi class multilabel audio classification challenges for animal sounds. However the setup (soundscapes vs xeno canto, weird evaluation metrics, different train/test labels) is completely different. What worked there might not work here…

", "votes": null, "replies": [] }, { "id": 1167289, "author_name": "yassinealouini", "author_url": "", "post_date": "01/24/2021 06:40:17", "content": "

Thanks a lot for the list. Indeed, some of the things have changed, I will keep that in mind. πŸ‘Œ

\n

I was planning to use librosa but now that you mention torchlibrosa, might be more suitable. πŸ‘

", "votes": null, "replies": [] } ] }, { "id": 1175993, "author_name": "tasneemabdulrahim", "author_url": "", "post_date": "01/29/2021 12:49:26", "content": "

Thanks for this list of links. Do you have an idea of most used features and models?

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1082388": "* **2020** https://www.kaggle.com/c/birdsong-recognition\n* **2019** https://www.kaggle.com/c/freesound-audio-tagging-2019\n* https://www.kaggle.com/c/the-icml-2013-bird-challenge\n* https://www.kaggle.com/c/mlsp-2013-birds\n* https://www.kaggle.com/c/whale-detection-challenge\n\n## Outside kaggle\n* 2020 https://www.aicrowd.com/challenges/lifeclef-2020-bird-monophone\n* 2019 https://www.crowdai.org/challenges/lifeclef-2019-bird-detection\n* 2018 http://dcase.community/challenge2018/task-bird-audio-detection", "1082390": "The recently finished **Cornell Birdcall Identification** top solutions collected by @addisonhoward \n- [1st Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183208)\n- [2nd Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183269)\n- [3rd Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183199)\n- [4th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183339)\n- [5th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183300)\n- [6th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183204)\n- [8th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183223)\n- [9th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183315)\n- [10th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183407)\n- [13th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183436)\n- [18th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183219)\n- [21st Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183209)\n- [36th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183222)\n- [39th Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183258)\n- [43rd Place Solution](https://www.kaggle.com/c/birdsong-recognition/discussion/183240)\n\n**Freesound Audio Tagging** 2019 solutions collected by @jesucristo \n- https://www.kaggle.com/c/birdsong-recognition/discussion/158943", "1082455": "Also good to see DCASE challenges:\nhttp://dcase.community/challenge2020/\nit holds multiple challenges every year and participants submit technical reports which are quite helpful to know state-of-the-art techniques.", "1165769": "Thanks for this list of links. Do you have an idea of most used features and models?", "1166911": "I did not checked the solutions in detail just had a glimpse at the descriptions. \nWhat I learned from the Cornell comp:\n* MelSpectrogram with torchlibrosa\n* SED\n* PANNs\n* Synthetic noise augmentation\n* External Noise augmentation \n\nBoth the Cornell and RFCX are multi class multilabel audio classification challenges for animal sounds. However the setup (soundscapes vs xeno canto, weird evaluation metrics, different train/test labels) is completely different. What worked there might not work here...", "1167289": "Thanks a lot for the list. Indeed, some of the things have changed, I will keep that in mind. πŸ‘Œ \n\nI was planning to use librosa but now that you mention torchlibrosa, might be more suitable. πŸ‘", "1175993": "Thanks for this list of links. Do you have an idea of most used features and models?" }, "source": "meta" }