{
  "id": 201377,
  "title": "RNN models",
  "url": "/competitions/rfcx-species-audio-detection/discussion/201377",
  "author_name": "",
  "post_date": "2020-12-04T15:42:29.945686600Z",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I came across this post on speech recognition using spectrogram fratures: <a href=\"https://www.assemblyai.com/blog/end-to-end-speech-recognition-pytorch\" target=\"_blank\">https://www.assemblyai.com/blog/end-to-end-speech-recognition-pytorch</a></p>\n<p>Has anyone tried similar approaches for this competition?</p>",
  "messages": [
    {
      "id": "1102110",
      "postDate": "12/04/2020 15:42:29",
      "content": "<p>I came across this post on speech recognition using spectrogram fratures: <a href=\"https://www.assemblyai.com/blog/end-to-end-speech-recognition-pytorch\" target=\"_blank\">https://www.assemblyai.com/blog/end-to-end-speech-recognition-pytorch</a></p>\n<p>Has anyone tried similar approaches for this competition?</p>",
      "rawMarkdown": "I came across this post on speech recognition using spectrogram fratures: https://www.assemblyai.com/blog/end-to-end-speech-recognition-pytorch\n\nHas anyone tried similar approaches for this competition?",
      "votes": null
    },
    {
      "id": "1102302",
      "postDate": "12/04/2020 19:11:19",
      "content": "<p>I'm in the middle of implementing parts of that approach. Specifically, I'm looking at Spectrograms with SpecAugment. However, the similarities end there. There's not a lot of labelled data, so I'm doing transfer learning off a common model.</p>",
      "rawMarkdown": "I'm in the middle of implementing parts of that approach. Specifically, I'm looking at Spectrograms with SpecAugment. However, the similarities end there. There's not a lot of labelled data, so I'm doing transfer learning off a common model.",
      "votes": null
    },
    {
      "id": "1122109",
      "postDate": "12/22/2020 07:40:21",
      "content": "<p>Taking too much time forget about it </p>",
      "rawMarkdown": "Taking too much time forget about it",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1102302,
      "author_name": "maltonji",
      "author_url": "",
      "post_date": "12/04/2020 19:11:19",
      "content": "<p>I'm in the middle of implementing parts of that approach. Specifically, I'm looking at Spectrograms with SpecAugment. However, the similarities end there. There's not a lot of labelled data, so I'm doing transfer learning off a common model.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1122109,
      "author_name": "swaralipibose",
      "author_url": "",
      "post_date": "12/22/2020 07:40:21",
      "content": "<p>Taking too much time forget about it </p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1102110": "I came across this post on speech recognition using spectrogram fratures: https://www.assemblyai.com/blog/end-to-end-speech-recognition-pytorch\n\nHas anyone tried similar approaches for this competition?",
    "1102302": "I'm in the middle of implementing parts of that approach. Specifically, I'm looking at Spectrograms with SpecAugment. However, the similarities end there. There's not a lot of labelled data, so I'm doing transfer learning off a common model.",
    "1122109": "Taking too much time forget about it"
  },
  "source": "meta"
}