{
  "id": 413040,
  "title": "Sound Event Detection (SED)?",
  "url": "/competitions/birdclef-2023/discussion/413040",
  "author_name": "Andy Atkinson",
  "post_date": "2023-05-26T13:34:59.423000",
  "votes": 4,
  "comment_count": 0,
  "views": 0,
  "content": "<p>There were quite a few mentions of SED in the top solutions.  What is SED? Should I be thinking of SED as a different type of vision backbone (attention transformer type?) instead of a CNN?  Or is there more too it, like is it more like a multi-stage approach?</p>\n<p>Here is an example cited, I can't seem to figure out how it works.<br>\n<a href=\"https://www.kaggle.com/code/hidehisaarai1213/pytorch-training-birdclef2021-starter/notebook\" target=\"_blank\">https://www.kaggle.com/code/hidehisaarai1213/pytorch-training-birdclef2021-starter/notebook</a></p>",
  "messages": [
    {
      "id": 2275086,
      "postDate": "2023-05-26T13:34:59.423Z",
      "content": "<p>There were quite a few mentions of SED in the top solutions.  What is SED? Should I be thinking of SED as a different type of vision backbone (attention transformer type?) instead of a CNN?  Or is there more too it, like is it more like a multi-stage approach?</p>\n<p>Here is an example cited, I can't seem to figure out how it works.<br>\n<a href=\"https://www.kaggle.com/code/hidehisaarai1213/pytorch-training-birdclef2021-starter/notebook\" target=\"_blank\">https://www.kaggle.com/code/hidehisaarai1213/pytorch-training-birdclef2021-starter/notebook</a></p>",
      "rawMarkdown": "There were quite a few mentions of SED in the top solutions.  What is SED? Should I be thinking of SED as a different type of vision backbone (attention transformer type?) instead of a CNN?  Or is there more too it, like is it more like a multi-stage approach?\n\nHere is an example cited, I can't seem to figure out how it works.\nhttps://www.kaggle.com/code/hidehisaarai1213/pytorch-training-birdclef2021-starter/notebook",
      "votes": 4
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "2275086": "There were quite a few mentions of SED in the top solutions.  What is SED? Should I be thinking of SED as a different type of vision backbone (attention transformer type?) instead of a CNN?  Or is there more too it, like is it more like a multi-stage approach?\n\nHere is an example cited, I can't seem to figure out how it works.\nhttps://www.kaggle.com/code/hidehisaarai1213/pytorch-training-birdclef2021-starter/notebook"
  }
}