{
  "id": 167264,
  "title": "Singing/Background Split Datasets",
  "url": "/competitions/birdsong-recognition/discussion/167264",
  "author_name": "",
  "post_date": "2020-07-15T20:21:02.475137900Z",
  "votes": 24,
  "comment_count": 7,
  "views": 0,
  "content": "<p>Following work of <a href=\"/radek1\">@radek1</a>  and <a href=\"/ttahara\">@ttahara</a>  , I have used their resample datasets to split clips between background noise and parts where the birds are singing.\nIntuition is developed in this notebook : <a href=\"https://www.kaggle.com/luisblanche/preprocessing-splitting-birds-sounds-background\">Preprocessing: splitting birds sounds/background</a></p>\n\n<p>There is too much data to process it on Kaggle so I took up a Google Cloud Engine to do it .[https://github.com/LuisBlanche/kaggle-birdsong-split-silence](Find code on github here)</p>\n\n<p>I am now uploading the datasets to kaggle, you can find them here : </p>\n\n<p><a href=\"https://www.kaggle.com/luisblanche/birdcall-singing-0\">https://www.kaggle.com/luisblanche/birdcall-singing-0\n</a></p>\n\n<p><a href=\"https://www.kaggle.com/luisblanche/birdcall-singing-1\">https://www.kaggle.com/luisblanche/birdcall-singing-1</a></p>\n\n<p><a href=\"https://www.kaggle.com/luisblanche/birdcall-singing-2\">https://www.kaggle.com/luisblanche/birdcall-singing-2</a></p>\n\n<p><a href=\"https://www.kaggle.com/luisblanche/birdcall-singing-3\">https://www.kaggle.com/luisblanche/birdcall-singing-3</a></p>\n\n<p><a href=\"https://www.kaggle.com/luisblanche/birdcall-singing-4\">https://www.kaggle.com/luisblanche/birdcall-singing-4</a></p>\n\n<p><a href=\"https://www.kaggle.com/luisblanche/birdcall-background\">https://www.kaggle.com/luisblanche/birdcall-background</a>  This one is 5sec samples of background sound for each clip of the training data </p>\n\n<p><strong>Edit</strong> : dataset number 4 is now public , do not hesitate to comment on this feed if you find that those datasets are useful for the challenge </p>",
  "messages": [
    {
      "id": "930929",
      "postDate": "07/15/2020 20:21:02",
      "content": "<p>Following work of <a href=\"/radek1\">@radek1</a>  and <a href=\"/ttahara\">@ttahara</a>  , I have used their resample datasets to split clips between background noise and parts where the birds are singing.\nIntuition is developed in this notebook : <a href=\"https://www.kaggle.com/luisblanche/preprocessing-splitting-birds-sounds-background\">Preprocessing: splitting birds sounds/background</a></p>\n\n<p>There is too much data to process it on Kaggle so I took up a Google Cloud Engine to do it .[https://github.com/LuisBlanche/kaggle-birdsong-split-silence](Find code on github here)</p>\n\n<p>I am now uploading the datasets to kaggle, you can find them here : </p>\n\n<p><a href=\"https://www.kaggle.com/luisblanche/birdcall-singing-0\">https://www.kaggle.com/luisblanche/birdcall-singing-0\n</a></p>\n\n<p><a href=\"https://www.kaggle.com/luisblanche/birdcall-singing-1\">https://www.kaggle.com/luisblanche/birdcall-singing-1</a></p>\n\n<p><a href=\"https://www.kaggle.com/luisblanche/birdcall-singing-2\">https://www.kaggle.com/luisblanche/birdcall-singing-2</a></p>\n\n<p><a href=\"https://www.kaggle.com/luisblanche/birdcall-singing-3\">https://www.kaggle.com/luisblanche/birdcall-singing-3</a></p>\n\n<p><a href=\"https://www.kaggle.com/luisblanche/birdcall-singing-4\">https://www.kaggle.com/luisblanche/birdcall-singing-4</a></p>\n\n<p><a href=\"https://www.kaggle.com/luisblanche/birdcall-background\">https://www.kaggle.com/luisblanche/birdcall-background</a>  This one is 5sec samples of background sound for each clip of the training data </p>\n\n<p><strong>Edit</strong> : dataset number 4 is now public , do not hesitate to comment on this feed if you find that those datasets are useful for the challenge </p>",
      "rawMarkdown": "Following work of @radek1  and @ttahara  , I have used their resample datasets to split clips between background noise and parts where the birds are singing.\nIntuition is developed in this notebook : [Preprocessing: splitting birds sounds/background](https://www.kaggle.com/luisblanche/preprocessing-splitting-birds-sounds-background)\n\nThere is too much data to process it on Kaggle so I took up a Google Cloud Engine to do it .[https://github.com/LuisBlanche/kaggle-birdsong-split-silence](Find code on github here)\n\nI am now uploading the datasets to kaggle, you can find them here : \n\n[https://www.kaggle.com/luisblanche/birdcall-singing-0\n](https://www.kaggle.com/luisblanche/birdcall-singing-0)\n\n[https://www.kaggle.com/luisblanche/birdcall-singing-1](https://www.kaggle.com/luisblanche/birdcall-singing-1)\n\n[https://www.kaggle.com/luisblanche/birdcall-singing-2](https://www.kaggle.com/luisblanche/birdcall-singing-2)\n\n[https://www.kaggle.com/luisblanche/birdcall-singing-3](https://www.kaggle.com/luisblanche/birdcall-singing-3)\n\n[https://www.kaggle.com/luisblanche/birdcall-singing-4](https://www.kaggle.com/luisblanche/birdcall-singing-4)\n\n[https://www.kaggle.com/luisblanche/birdcall-background](https://www.kaggle.com/luisblanche/birdcall-background)  This one is 5sec samples of background sound for each clip of the training data \n\n**Edit** : dataset number 4 is now public , do not hesitate to comment on this feed if you find that those datasets are useful for the challenge",
      "votes": null
    },
    {
      "id": "930941",
      "postDate": "07/15/2020 20:31:21",
      "content": "<p>Great work, Thanks!</p>\n\n<p>I have a question. Is the splitting method the same as <a href=\"https://www.kaggle.com/luisblanche/preprocessing-splitting-birds-sounds-background\">your notebook</a>? (I'm trying it in my local environment.)</p>",
      "rawMarkdown": "Great work, Thanks!\n\nI have a question. Is the splitting method the same as [your notebook](https://www.kaggle.com/luisblanche/preprocessing-splitting-birds-sounds-background)? (I'm trying it in my local environment.)",
      "votes": null
    },
    {
      "id": "930979",
      "postDate": "07/15/2020 21:22:31",
      "content": "<p>Yes it is the same. You can find the code I used on my github here : <a href=\"https://github.com/LuisBlanche/kaggle-birdsong-split-silence\">https://github.com/LuisBlanche/kaggle-birdsong-split-silence</a></p>\n\n<p>It took about 1h to split everything with an 8 core Google Cloud Compute Engine</p>",
      "rawMarkdown": "Yes it is the same. You can find the code I used on my github here : [https://github.com/LuisBlanche/kaggle-birdsong-split-silence](https://github.com/LuisBlanche/kaggle-birdsong-split-silence)\n\nIt took about 1h to split everything with an 8 core Google Cloud Compute Engine",
      "votes": null
    },
    {
      "id": "931375",
      "postDate": "07/16/2020 06:55:03",
      "content": "<p>how amazing👌 </p>",
      "rawMarkdown": "how amazing👌",
      "votes": null
    },
    {
      "id": "931390",
      "postDate": "07/16/2020 07:03:00",
      "content": "<p>Thanks!</p>",
      "rawMarkdown": "Thanks!",
      "votes": null
    },
    {
      "id": "943115",
      "postDate": "07/24/2020 06:33:05",
      "content": "<p><a href=\"/luisblanche\">@luisblanche</a> Can you please post the below dataset again\n<code>https://www.kaggle.com/luisblanche/birdcall-singing-4. (Still uploading)</code>\nCurrent one is giving error\nThanks</p>",
      "rawMarkdown": "luisblanche Can you please post the below dataset again\n`https://www.kaggle.com/luisblanche/birdcall-singing-4. (Still uploading)`\nCurrent one is giving error\nThanks",
      "votes": null
    },
    {
      "id": "944965",
      "postDate": "07/25/2020 13:30:59",
      "content": "<p>Hi yes I need to make it public, I'll do or sometime next week</p>",
      "rawMarkdown": "Hi yes I need to make it public, I'll do or sometime next week",
      "votes": null
    },
    {
      "id": "945042",
      "postDate": "07/25/2020 14:32:05",
      "content": "<p>Thanks, Inspired by your work I have also implemented somewhat similar approach <a href=\"https://www.kaggle.com/jainarindam/imp-remove-background-dead-noise\">here</a> .\nI hope that can further make audio quality even better.</p>",
      "rawMarkdown": "Thanks, Inspired by your work I have also implemented somewhat similar approach [here](https://www.kaggle.com/jainarindam/imp-remove-background-dead-noise) .\nI hope that can further make audio quality even better.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 930941,
      "author_name": "ttahara",
      "author_url": "",
      "post_date": "07/15/2020 20:31:21",
      "content": "<p>Great work, Thanks!</p>\n\n<p>I have a question. Is the splitting method the same as <a href=\"https://www.kaggle.com/luisblanche/preprocessing-splitting-birds-sounds-background\">your notebook</a>? (I'm trying it in my local environment.)</p>",
      "votes": null,
      "replies": [
        {
          "id": 930979,
          "author_name": "luisblanche",
          "author_url": "",
          "post_date": "07/15/2020 21:22:31",
          "content": "<p>Yes it is the same. You can find the code I used on my github here : <a href=\"https://github.com/LuisBlanche/kaggle-birdsong-split-silence\">https://github.com/LuisBlanche/kaggle-birdsong-split-silence</a></p>\n\n<p>It took about 1h to split everything with an 8 core Google Cloud Compute Engine</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 931390,
          "author_name": "ttahara",
          "author_url": "",
          "post_date": "07/16/2020 07:03:00",
          "content": "<p>Thanks!</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 931375,
      "author_name": "leila73",
      "author_url": "",
      "post_date": "07/16/2020 06:55:03",
      "content": "<p>how amazing👌 </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 943115,
      "author_name": "jainarindam",
      "author_url": "",
      "post_date": "07/24/2020 06:33:05",
      "content": "<p><a href=\"/luisblanche\">@luisblanche</a> Can you please post the below dataset again\n<code>https://www.kaggle.com/luisblanche/birdcall-singing-4. (Still uploading)</code>\nCurrent one is giving error\nThanks</p>",
      "votes": null,
      "replies": [
        {
          "id": 944965,
          "author_name": "luisblanche",
          "author_url": "",
          "post_date": "07/25/2020 13:30:59",
          "content": "<p>Hi yes I need to make it public, I'll do or sometime next week</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 945042,
          "author_name": "jainarindam",
          "author_url": "",
          "post_date": "07/25/2020 14:32:05",
          "content": "<p>Thanks, Inspired by your work I have also implemented somewhat similar approach <a href=\"https://www.kaggle.com/jainarindam/imp-remove-background-dead-noise\">here</a> .\nI hope that can further make audio quality even better.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "930929": "Following work of @radek1  and @ttahara  , I have used their resample datasets to split clips between background noise and parts where the birds are singing.\nIntuition is developed in this notebook : [Preprocessing: splitting birds sounds/background](https://www.kaggle.com/luisblanche/preprocessing-splitting-birds-sounds-background)\n\nThere is too much data to process it on Kaggle so I took up a Google Cloud Engine to do it .[https://github.com/LuisBlanche/kaggle-birdsong-split-silence](Find code on github here)\n\nI am now uploading the datasets to kaggle, you can find them here : \n\n[https://www.kaggle.com/luisblanche/birdcall-singing-0\n](https://www.kaggle.com/luisblanche/birdcall-singing-0)\n\n[https://www.kaggle.com/luisblanche/birdcall-singing-1](https://www.kaggle.com/luisblanche/birdcall-singing-1)\n\n[https://www.kaggle.com/luisblanche/birdcall-singing-2](https://www.kaggle.com/luisblanche/birdcall-singing-2)\n\n[https://www.kaggle.com/luisblanche/birdcall-singing-3](https://www.kaggle.com/luisblanche/birdcall-singing-3)\n\n[https://www.kaggle.com/luisblanche/birdcall-singing-4](https://www.kaggle.com/luisblanche/birdcall-singing-4)\n\n[https://www.kaggle.com/luisblanche/birdcall-background](https://www.kaggle.com/luisblanche/birdcall-background)  This one is 5sec samples of background sound for each clip of the training data \n\n**Edit** : dataset number 4 is now public , do not hesitate to comment on this feed if you find that those datasets are useful for the challenge",
    "930941": "Great work, Thanks!\n\nI have a question. Is the splitting method the same as [your notebook](https://www.kaggle.com/luisblanche/preprocessing-splitting-birds-sounds-background)? (I'm trying it in my local environment.)",
    "930979": "Yes it is the same. You can find the code I used on my github here : [https://github.com/LuisBlanche/kaggle-birdsong-split-silence](https://github.com/LuisBlanche/kaggle-birdsong-split-silence)\n\nIt took about 1h to split everything with an 8 core Google Cloud Compute Engine",
    "931375": "how amazing👌",
    "931390": "Thanks!",
    "943115": "luisblanche Can you please post the below dataset again\n`https://www.kaggle.com/luisblanche/birdcall-singing-4. (Still uploading)`\nCurrent one is giving error\nThanks",
    "944965": "Hi yes I need to make it public, I'll do or sometime next week",
    "945042": "Thanks, Inspired by your work I have also implemented somewhat similar approach [here](https://www.kaggle.com/jainarindam/imp-remove-background-dead-noise) .\nI hope that can further make audio quality even better."
  },
  "source": "meta"
}