{
  "id": 45364,
  "title": "Google Speech Recognition Tutorial",
  "url": "/competitions/tensorflow-speech-recognition-challenge/discussion/45364",
  "author_name": "",
  "post_date": "2017-12-10T01:07:55.881261500Z",
  "votes": 1,
  "comment_count": 4,
  "views": 0,
  "content": "<p>I'm a bit confused by the fact that there's a tutorial on TensorFlow using the same data set. What score does that model get? Are people using that as a starting point?</p>",
  "messages": [
    {
      "id": "255754",
      "postDate": "12/10/2017 01:07:55",
      "content": "<p>I'm a bit confused by the fact that there's a tutorial on TensorFlow using the same data set. What score does that model get? Are people using that as a starting point?</p>",
      "rawMarkdown": "I'm a bit confused by the fact that there's a tutorial on TensorFlow using the same data set. What score does that model get? Are people using that as a starting point?",
      "votes": null
    },
    {
      "id": "255759",
      "postDate": "12/10/2017 01:25:23",
      "content": "<p>see <a href=\"https://www.kaggle.com/c/tensorflow-speech-recognition-challenge/discussion/44250\">https://www.kaggle.com/c/tensorflow-speech-recognition-challenge/discussion/44250</a></p>",
      "rawMarkdown": "see https://www.kaggle.com/c/tensorflow-speech-recognition-challenge/discussion/44250",
      "votes": null
    },
    {
      "id": "255765",
      "postDate": "12/10/2017 01:42:40",
      "content": "<p>Thanks Ren, I see they're saying it yields 0.77</p>",
      "rawMarkdown": "Thanks Ren, I see they're saying it yields 0.77",
      "votes": null
    },
    {
      "id": "255773",
      "postDate": "12/10/2017 03:01:29",
      "content": "<p>Score on Kaggle: 0.78</p>\n\n<p>Yes, it can be used as a starting point. The examples uses MFCC spectogram as inputs features and presents how to use convolutions. So it is a great starting point.</p>",
      "rawMarkdown": "Score on Kaggle: 0.78\n\nYes, it can be used as a starting point. The examples uses MFCC spectogram as inputs features and presents how to use convolutions. So it is a great starting point.",
      "votes": null
    },
    {
      "id": "255923",
      "postDate": "12/10/2017 15:18:51",
      "content": "<p>Yes, 0.77 is what I got.</p>",
      "rawMarkdown": "Yes, 0.77 is what I got.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 255759,
      "author_name": "ryanzhang",
      "author_url": "",
      "post_date": "12/10/2017 01:25:23",
      "content": "<p>see <a href=\"https://www.kaggle.com/c/tensorflow-speech-recognition-challenge/discussion/44250\">https://www.kaggle.com/c/tensorflow-speech-recognition-challenge/discussion/44250</a></p>",
      "votes": null,
      "replies": [
        {
          "id": 255765,
          "author_name": "nscozzaro",
          "author_url": "",
          "post_date": "12/10/2017 01:42:40",
          "content": "<p>Thanks Ren, I see they're saying it yields 0.77</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 255923,
          "author_name": "shujian",
          "author_url": "",
          "post_date": "12/10/2017 15:18:51",
          "content": "<p>Yes, 0.77 is what I got.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 255773,
      "author_name": "barbolo",
      "author_url": "",
      "post_date": "12/10/2017 03:01:29",
      "content": "<p>Score on Kaggle: 0.78</p>\n\n<p>Yes, it can be used as a starting point. The examples uses MFCC spectogram as inputs features and presents how to use convolutions. So it is a great starting point.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "255754": "I'm a bit confused by the fact that there's a tutorial on TensorFlow using the same data set. What score does that model get? Are people using that as a starting point?",
    "255759": "see https://www.kaggle.com/c/tensorflow-speech-recognition-challenge/discussion/44250",
    "255765": "Thanks Ren, I see they're saying it yields 0.77",
    "255773": "Score on Kaggle: 0.78\n\nYes, it can be used as a starting point. The examples uses MFCC spectogram as inputs features and presents how to use convolutions. So it is a great starting point.",
    "255923": "Yes, 0.77 is what I got."
  },
  "source": "meta"
}