{
  "id": 121373,
  "title": "Docker ffmpeg request",
  "url": "/competitions/deepfake-detection-challenge/discussion/121373",
  "author_name": "",
  "post_date": "2019-12-12T18:45:29.206206200Z",
  "votes": 12,
  "comment_count": 5,
  "views": 0,
  "content": "<p>As far as I know, there is no way to extract an audio stream from mp4. Administrators, could you install ffmpeg? Or maybe someone knows how to do this without the Internet?</p>",
  "messages": [
    {
      "id": "693751",
      "postDate": "12/12/2019 18:45:29",
      "content": "<p>As far as I know, there is no way to extract an audio stream from mp4. Administrators, could you install ffmpeg? Or maybe someone knows how to do this without the Internet?</p>",
      "rawMarkdown": "As far as I know, there is no way to extract an audio stream from mp4. Administrators, could you install ffmpeg? Or maybe someone knows how to do this without the Internet?",
      "votes": null
    },
    {
      "id": "693782",
      "postDate": "12/12/2019 19:36:18",
      "content": "<p>Been looking around too and ffmpeg seems like the way to go.</p>",
      "rawMarkdown": "Been looking around too and ffmpeg seems like the way to go.",
      "votes": null
    },
    {
      "id": "694430",
      "postDate": "12/13/2019 15:32:53",
      "content": "<p>I have just posted a kernel that might help <a href=\"https://www.kaggle.com/rakibilly/extract-audio-starter\">https://www.kaggle.com/rakibilly/extract-audio-starter</a> </p>",
      "rawMarkdown": "I have just posted a kernel that might help https://www.kaggle.com/rakibilly/extract-audio-starter",
      "votes": null
    },
    {
      "id": "694444",
      "postDate": "12/13/2019 15:58:20",
      "content": "<p>Thank you! I hope it's legal)</p>",
      "rawMarkdown": "Thank you! I hope it's legal)",
      "votes": null
    },
    {
      "id": "695403",
      "postDate": "12/15/2019 04:44:39",
      "content": "<p>Would librosa work for you instead?</p>\n\n<p><code>\nimport librosa\ny,sr=librosa.load(\"some_file.mp4\")\n</code></p>\n\n<p>On my own computer, this gives me an ndarray y which looks roughly like a waveform.  I haven't tried this on kaggle's docker images yet, but if I'm looking at the right dockerfile, librosa does appear there.</p>",
      "rawMarkdown": "Would librosa work for you instead?\n\n```\nimport librosa\ny,sr=librosa.load(\"some_file.mp4\")\n```\n\nOn my own computer, this gives me an ndarray y which looks roughly like a waveform.  I haven't tried this on kaggle's docker images yet, but if I'm looking at the right dockerfile, librosa does appear there.",
      "votes": null
    },
    {
      "id": "695479",
      "postDate": "12/15/2019 08:02:19",
      "content": "<p>No, that didn't work either. Librosa usually uses ffmpeg as backend.</p>",
      "rawMarkdown": "No, that didn't work either. Librosa usually uses ffmpeg as backend.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 693782,
      "author_name": "jesperdramsch",
      "author_url": "",
      "post_date": "12/12/2019 19:36:18",
      "content": "<p>Been looking around too and ffmpeg seems like the way to go.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 694430,
      "author_name": "rakibilly",
      "author_url": "",
      "post_date": "12/13/2019 15:32:53",
      "content": "<p>I have just posted a kernel that might help <a href=\"https://www.kaggle.com/rakibilly/extract-audio-starter\">https://www.kaggle.com/rakibilly/extract-audio-starter</a> </p>",
      "votes": null,
      "replies": [
        {
          "id": 694444,
          "author_name": "sorokin",
          "author_url": "",
          "post_date": "12/13/2019 15:58:20",
          "content": "<p>Thank you! I hope it's legal)</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 695403,
      "author_name": "particlebbq",
      "author_url": "",
      "post_date": "12/15/2019 04:44:39",
      "content": "<p>Would librosa work for you instead?</p>\n\n<p><code>\nimport librosa\ny,sr=librosa.load(\"some_file.mp4\")\n</code></p>\n\n<p>On my own computer, this gives me an ndarray y which looks roughly like a waveform.  I haven't tried this on kaggle's docker images yet, but if I'm looking at the right dockerfile, librosa does appear there.</p>",
      "votes": null,
      "replies": [
        {
          "id": 695479,
          "author_name": "sorokin",
          "author_url": "",
          "post_date": "12/15/2019 08:02:19",
          "content": "<p>No, that didn't work either. Librosa usually uses ffmpeg as backend.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "693751": "As far as I know, there is no way to extract an audio stream from mp4. Administrators, could you install ffmpeg? Or maybe someone knows how to do this without the Internet?",
    "693782": "Been looking around too and ffmpeg seems like the way to go.",
    "694430": "I have just posted a kernel that might help https://www.kaggle.com/rakibilly/extract-audio-starter",
    "694444": "Thank you! I hope it's legal)",
    "695403": "Would librosa work for you instead?\n\n```\nimport librosa\ny,sr=librosa.load(\"some_file.mp4\")\n```\n\nOn my own computer, this gives me an ndarray y which looks roughly like a waveform.  I haven't tried this on kaggle's docker images yet, but if I'm looking at the right dockerfile, librosa does appear there.",
    "695479": "No, that didn't work either. Librosa usually uses ffmpeg as backend."
  },
  "source": "meta"
}