{
  "id": 307879,
  "title": "Pre-trained Transformer models for audio",
  "url": "/competitions/birdclef-2022/discussion/307879",
  "author_name": "",
  "post_date": "2022-02-16T03:24:30.529732600Z",
  "votes": 45,
  "comment_count": 4,
  "views": 0,
  "content": "<p>I show you <strong>pre-trained</strong> Transformer models for audio.</p>\n<ul>\n<li><p><a href=\"https://arxiv.org/abs/2104.01778\" target=\"_blank\">AST: Audio Spectrogram Transformer</a><br>\ngithub: <a href=\"https://github.com/YuanGongND/ast\" target=\"_blank\">https://github.com/YuanGongND/ast</a></p></li>\n<li><p><a href=\"https://arxiv.org/abs/2110.05069\" target=\"_blank\">PaSST</a><br>\ngithub: <a href=\"https://github.com/kkoutini/PaSST\" target=\"_blank\">https://github.com/kkoutini/PaSST</a></p></li>\n</ul>\n<p>Attention! If you use AST, <strong>do not use librosa</strong>. You should always use torchaudio. I could not train AST well with this mistake.</p>",
  "messages": [
    {
      "id": "1692431",
      "postDate": "02/16/2022 03:24:30",
      "content": "<p>I show you <strong>pre-trained</strong> Transformer models for audio.</p>\n<ul>\n<li><p><a href=\"https://arxiv.org/abs/2104.01778\" target=\"_blank\">AST: Audio Spectrogram Transformer</a><br>\ngithub: <a href=\"https://github.com/YuanGongND/ast\" target=\"_blank\">https://github.com/YuanGongND/ast</a></p></li>\n<li><p><a href=\"https://arxiv.org/abs/2110.05069\" target=\"_blank\">PaSST</a><br>\ngithub: <a href=\"https://github.com/kkoutini/PaSST\" target=\"_blank\">https://github.com/kkoutini/PaSST</a></p></li>\n</ul>\n<p>Attention! If you use AST, <strong>do not use librosa</strong>. You should always use torchaudio. I could not train AST well with this mistake.</p>",
      "rawMarkdown": "I show you **pre-trained** Transformer models for audio.\n\n- [AST: Audio Spectrogram Transformer](https://arxiv.org/abs/2104.01778)\n github: https://github.com/YuanGongND/ast\n\n- [PaSST](https://arxiv.org/abs/2110.05069)\n github: https://github.com/kkoutini/PaSST\n\nAttention! If you use AST, **do not use librosa**. You should always use torchaudio. I could not train AST well with this mistake.",
      "votes": null
    },
    {
      "id": "1692664",
      "postDate": "02/16/2022 07:02:49",
      "content": "<p>Great, Thanks for sharing!!!</p>",
      "rawMarkdown": "Great, Thanks for sharing!!!",
      "votes": null
    },
    {
      "id": "1704285",
      "postDate": "02/25/2022 11:27:34",
      "content": "<p>Thanks, I will take a look</p>",
      "rawMarkdown": "Thanks, I will take a look",
      "votes": null
    },
    {
      "id": "1772254",
      "postDate": "04/30/2022 04:17:15",
      "content": "<p>thanks for sharing</p>",
      "rawMarkdown": "thanks for sharing",
      "votes": null
    },
    {
      "id": "1789736",
      "postDate": "05/14/2022 07:15:23",
      "content": "<p>Did you use transformers?<br>\n<a href=\"https://www.kaggle.com/shinmurashinmura\" target=\"_blank\">@shinmurashinmura</a> is transformers are working well?</p>",
      "rawMarkdown": "Did you use transformers?\n@shinmurashinmura is transformers are working well?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1692664,
      "author_name": "lallucycle",
      "author_url": "",
      "post_date": "02/16/2022 07:02:49",
      "content": "<p>Great, Thanks for sharing!!!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1704285,
      "author_name": "paulojunqueira",
      "author_url": "",
      "post_date": "02/25/2022 11:27:34",
      "content": "<p>Thanks, I will take a look</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1772254,
      "author_name": "rjr5357",
      "author_url": "",
      "post_date": "04/30/2022 04:17:15",
      "content": "<p>thanks for sharing</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1789736,
      "author_name": "vatsalmavani",
      "author_url": "",
      "post_date": "05/14/2022 07:15:23",
      "content": "<p>Did you use transformers?<br>\n<a href=\"https://www.kaggle.com/shinmurashinmura\" target=\"_blank\">@shinmurashinmura</a> is transformers are working well?</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1692431": "I show you **pre-trained** Transformer models for audio.\n\n- [AST: Audio Spectrogram Transformer](https://arxiv.org/abs/2104.01778)\n github: https://github.com/YuanGongND/ast\n\n- [PaSST](https://arxiv.org/abs/2110.05069)\n github: https://github.com/kkoutini/PaSST\n\nAttention! If you use AST, **do not use librosa**. You should always use torchaudio. I could not train AST well with this mistake.",
    "1692664": "Great, Thanks for sharing!!!",
    "1704285": "Thanks, I will take a look",
    "1772254": "thanks for sharing",
    "1789736": "Did you use transformers?\n@shinmurashinmura is transformers are working well?"
  },
  "source": "meta"
}