{
  "id": 44017,
  "title": "File train/audio/bird/3e7124ba_nohash_0.wav is silent",
  "url": "/competitions/tensorflow-speech-recognition-challenge/discussion/44017",
  "author_name": "",
  "post_date": "2017-11-22T10:37:04.633159400Z",
  "votes": 3,
  "comment_count": 4,
  "views": 0,
  "content": "<p>Hello,\nthe file <code>train/audio/bird/3e7124ba_nohash_0.wav</code> seem full of silence ! \nIt has the same length as others but no sound.\nIs it a problem on my side ? </p>\n\n<pre><code>ls -lsa /media/Red1000/Datasets/TFSRC/train/audio/bird/3e*\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e2ba5f7_nohash_0.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e31dffe_nohash_0.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e3b65eb_nohash_0.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e549219_nohash_0.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e549219_nohash_1.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e7124ba_nohash_0.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3ec05c3d_nohash_0.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3efa7ec4_nohash_0.wav\n</code></pre>",
  "messages": [
    {
      "id": "247102",
      "postDate": "11/22/2017 10:37:04",
      "content": "<p>Hello,\nthe file <code>train/audio/bird/3e7124ba_nohash_0.wav</code> seem full of silence ! \nIt has the same length as others but no sound.\nIs it a problem on my side ? </p>\n\n<pre><code>ls -lsa /media/Red1000/Datasets/TFSRC/train/audio/bird/3e*\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e2ba5f7_nohash_0.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e31dffe_nohash_0.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e3b65eb_nohash_0.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e549219_nohash_0.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e549219_nohash_1.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e7124ba_nohash_0.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3ec05c3d_nohash_0.wav\n32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3efa7ec4_nohash_0.wav\n</code></pre>",
      "rawMarkdown": "Hello,\nthe file `train/audio/bird/3e7124ba_nohash_0.wav` seem full of silence ! \nIt has the same length as others but no sound.\nIs it a problem on my side ? \n\n    ls -lsa /media/Red1000/Datasets/TFSRC/train/audio/bird/3e*\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e2ba5f7_nohash_0.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e31dffe_nohash_0.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e3b65eb_nohash_0.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e549219_nohash_0.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e549219_nohash_1.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e7124ba_nohash_0.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3ec05c3d_nohash_0.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3efa7ec4_nohash_0.wav",
      "votes": null
    },
    {
      "id": "247124",
      "postDate": "11/22/2017 11:57:52",
      "content": "<p>No, it's all zeros after loading here, too. (I just noticed and wanted to check the forums, but you beat me to asking.)</p>\n\n<p>I hope the bird is doing well. (No animals were harmed for this cheap joke.)</p>",
      "rawMarkdown": "No, it's all zeros after loading here, too. (I just noticed and wanted to check the forums, but you beat me to asking.)\n\nI hope the bird is doing well. (No animals were harmed for this cheap joke.)",
      "votes": null
    },
    {
      "id": "248264",
      "postDate": "11/25/2017 12:40:00",
      "content": "<p>the bird in train/audio/tree/611d2b50_nohash_0.wav and train/audio/three/611d2b50_nohash_4.wav seems to be doing well</p>",
      "rawMarkdown": "the bird in train/audio/tree/611d2b50_nohash_0.wav and train/audio/three/611d2b50_nohash_4.wav seems to be doing well",
      "votes": null
    },
    {
      "id": "248379",
      "postDate": "11/25/2017 22:16:28",
      "content": "<p>I went through around 470 of the 'least amplitude' sounds (according to the difference of the maximum and minimum sample), see this kernel: <a href=\"https://www.kaggle.com/holzner/almost-empty-samples/notebook\">https://www.kaggle.com/holzner/almost-empty-samples/notebook</a></p>\n\n<p>Most of them are noise even though they have a label associated. </p>\n\n<p>Some noise sounds in this set are associated to the same speaker but on the other hand, some other sounds (beyond the 470 sounds) of the same speaker are clearly understandable. From around the 462th sound on (in the train set) the sounds become human understandable.</p>",
      "rawMarkdown": "I went through around 470 of the 'least amplitude' sounds (according to the difference of the maximum and minimum sample), see this kernel: https://www.kaggle.com/holzner/almost-empty-samples/notebook\n\nMost of them are noise even though they have a label associated. \n\nSome noise sounds in this set are associated to the same speaker but on the other hand, some other sounds (beyond the 470 sounds) of the same speaker are clearly understandable. From around the 462th sound on (in the train set) the sounds become human understandable.",
      "votes": null
    },
    {
      "id": "248825",
      "postDate": "11/27/2017 02:40:22",
      "content": "<p>There seem to be quite a lot of misclassified samples in the training set... I was looking through them and found wrong words even. I will bother to write them down next time...</p>",
      "rawMarkdown": "There seem to be quite a lot of misclassified samples in the training set... I was looking through them and found wrong words even. I will bother to write them down next time...",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 247124,
      "author_name": "tviehmann",
      "author_url": "",
      "post_date": "11/22/2017 11:57:52",
      "content": "<p>No, it's all zeros after loading here, too. (I just noticed and wanted to check the forums, but you beat me to asking.)</p>\n\n<p>I hope the bird is doing well. (No animals were harmed for this cheap joke.)</p>",
      "votes": null,
      "replies": [
        {
          "id": 248264,
          "author_name": "holzner",
          "author_url": "",
          "post_date": "11/25/2017 12:40:00",
          "content": "<p>the bird in train/audio/tree/611d2b50_nohash_0.wav and train/audio/three/611d2b50_nohash_4.wav seems to be doing well</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 248379,
      "author_name": "holzner",
      "author_url": "",
      "post_date": "11/25/2017 22:16:28",
      "content": "<p>I went through around 470 of the 'least amplitude' sounds (according to the difference of the maximum and minimum sample), see this kernel: <a href=\"https://www.kaggle.com/holzner/almost-empty-samples/notebook\">https://www.kaggle.com/holzner/almost-empty-samples/notebook</a></p>\n\n<p>Most of them are noise even though they have a label associated. </p>\n\n<p>Some noise sounds in this set are associated to the same speaker but on the other hand, some other sounds (beyond the 470 sounds) of the same speaker are clearly understandable. From around the 462th sound on (in the train set) the sounds become human understandable.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 248825,
      "author_name": "douglasandrade",
      "author_url": "",
      "post_date": "11/27/2017 02:40:22",
      "content": "<p>There seem to be quite a lot of misclassified samples in the training set... I was looking through them and found wrong words even. I will bother to write them down next time...</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "247102": "Hello,\nthe file `train/audio/bird/3e7124ba_nohash_0.wav` seem full of silence ! \nIt has the same length as others but no sound.\nIs it a problem on my side ? \n\n    ls -lsa /media/Red1000/Datasets/TFSRC/train/audio/bird/3e*\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e2ba5f7_nohash_0.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e31dffe_nohash_0.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e3b65eb_nohash_0.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e549219_nohash_0.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e549219_nohash_1.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3e7124ba_nohash_0.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3ec05c3d_nohash_0.wav\n    32 -rw-r----- 1 tl tl 32044 Nov 13 21:11 /media/Red1000/Datasets/TFSRC/train/audio/bird/3efa7ec4_nohash_0.wav",
    "247124": "No, it's all zeros after loading here, too. (I just noticed and wanted to check the forums, but you beat me to asking.)\n\nI hope the bird is doing well. (No animals were harmed for this cheap joke.)",
    "248264": "the bird in train/audio/tree/611d2b50_nohash_0.wav and train/audio/three/611d2b50_nohash_4.wav seems to be doing well",
    "248379": "I went through around 470 of the 'least amplitude' sounds (according to the difference of the maximum and minimum sample), see this kernel: https://www.kaggle.com/holzner/almost-empty-samples/notebook\n\nMost of them are noise even though they have a label associated. \n\nSome noise sounds in this set are associated to the same speaker but on the other hand, some other sounds (beyond the 470 sounds) of the same speaker are clearly understandable. From around the 462th sound on (in the train set) the sounds become human understandable.",
    "248825": "There seem to be quite a lot of misclassified samples in the training set... I was looking through them and found wrong words even. I will bother to write them down next time..."
  },
  "source": "meta"
}