{
  "id": 440179,
  "title": "Wav2Vec Research papers and articles",
  "url": "/competitions/bengaliai-speech/discussion/440179",
  "author_name": "",
  "post_date": "2023-09-14T04:02:42.308723600Z",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>With respect to Bengali.AI Speech Recognition Competition, I observe the important model which <br>\nis useful is Wav2Vec model.  I am sharing here the research papers which are relevant for the participants:</p>\n<ol>\n<li>WAV2VEC: UNSUPERVISED PRE-TRAINING FOR SPEECH RECOGNITION<br>\n<a href=\"https://arxiv.org/pdf/1904.05862.pdf\" target=\"_blank\">https://arxiv.org/pdf/1904.05862.pdf</a></li>\n<li>Wav2Vec Model doc: <a href=\"https://huggingface.co/docs/transformers/model_doc/wav2vec2\" target=\"_blank\">https://huggingface.co/docs/transformers/model_doc/wav2vec2</a></li>\n<li>An Illustrated Tour of Wav2Vec : <a href=\"https://jonathanbgn.com/2021/09/30/illustrated-wav2vec-2.html\" target=\"_blank\">https://jonathanbgn.com/2021/09/30/illustrated-wav2vec-2.html</a></li>\n<li>wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations<br>\n<a href=\"https://proceedings.neurips.cc/paper/2020/file/92d1e1eb1cd6f9fba3227870bb6d7f07-Paper.pdf\" target=\"_blank\">https://proceedings.neurips.cc/paper/2020/file/92d1e1eb1cd6f9fba3227870bb6d7f07-Paper.pdf</a></li>\n<li>Speech to Text with Wav2Vec: <a href=\"https://www.kdnuggets.com/2021/03/speech-text-wav2vec.html\" target=\"_blank\">https://www.kdnuggets.com/2021/03/speech-text-wav2vec.html</a></li>\n<li>Automatic Speech Recognition wth Wav2Vec: <a href=\"https://www.analyticsvidhya.com/blog/2022/06/automatic-speech-recognition-using-wav2vec2/\" target=\"_blank\">https://www.analyticsvidhya.com/blog/2022/06/automatic-speech-recognition-using-wav2vec2/</a></li>\n</ol>",
  "messages": [
    {
      "id": "2438100",
      "postDate": "09/14/2023 04:02:42",
      "content": "<p>With respect to Bengali.AI Speech Recognition Competition, I observe the important model which <br>\nis useful is Wav2Vec model.  I am sharing here the research papers which are relevant for the participants:</p>\n<ol>\n<li>WAV2VEC: UNSUPERVISED PRE-TRAINING FOR SPEECH RECOGNITION<br>\n<a href=\"https://arxiv.org/pdf/1904.05862.pdf\" target=\"_blank\">https://arxiv.org/pdf/1904.05862.pdf</a></li>\n<li>Wav2Vec Model doc: <a href=\"https://huggingface.co/docs/transformers/model_doc/wav2vec2\" target=\"_blank\">https://huggingface.co/docs/transformers/model_doc/wav2vec2</a></li>\n<li>An Illustrated Tour of Wav2Vec : <a href=\"https://jonathanbgn.com/2021/09/30/illustrated-wav2vec-2.html\" target=\"_blank\">https://jonathanbgn.com/2021/09/30/illustrated-wav2vec-2.html</a></li>\n<li>wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations<br>\n<a href=\"https://proceedings.neurips.cc/paper/2020/file/92d1e1eb1cd6f9fba3227870bb6d7f07-Paper.pdf\" target=\"_blank\">https://proceedings.neurips.cc/paper/2020/file/92d1e1eb1cd6f9fba3227870bb6d7f07-Paper.pdf</a></li>\n<li>Speech to Text with Wav2Vec: <a href=\"https://www.kdnuggets.com/2021/03/speech-text-wav2vec.html\" target=\"_blank\">https://www.kdnuggets.com/2021/03/speech-text-wav2vec.html</a></li>\n<li>Automatic Speech Recognition wth Wav2Vec: <a href=\"https://www.analyticsvidhya.com/blog/2022/06/automatic-speech-recognition-using-wav2vec2/\" target=\"_blank\">https://www.analyticsvidhya.com/blog/2022/06/automatic-speech-recognition-using-wav2vec2/</a></li>\n</ol>",
      "rawMarkdown": "With respect to Bengali.AI Speech Recognition Competition, I observe the important model which \nis useful is Wav2Vec model.  I am sharing here the research papers which are relevant for the participants:\n\n1. WAV2VEC: UNSUPERVISED PRE-TRAINING FOR SPEECH RECOGNITION\nhttps://arxiv.org/pdf/1904.05862.pdf\n2.  Wav2Vec Model doc: https://huggingface.co/docs/transformers/model_doc/wav2vec2\n3.  An Illustrated Tour of Wav2Vec : https://jonathanbgn.com/2021/09/30/illustrated-wav2vec-2.html\n4. wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations\nhttps://proceedings.neurips.cc/paper/2020/file/92d1e1eb1cd6f9fba3227870bb6d7f07-Paper.pdf\n5. Speech to Text with Wav2Vec: https://www.kdnuggets.com/2021/03/speech-text-wav2vec.html\n6. Automatic Speech Recognition wth Wav2Vec: https://www.analyticsvidhya.com/blog/2022/06/automatic-speech-recognition-using-wav2vec2/",
      "votes": null
    },
    {
      "id": "2438119",
      "postDate": "09/14/2023 04:31:59",
      "content": "<p>Great resources <a href=\"https://www.kaggle.com/crsuthikshnkumar\" target=\"_blank\">@crsuthikshnkumar</a>, thanks for sharing.</p>",
      "rawMarkdown": "Great resources @crsuthikshnkumar, thanks for sharing.",
      "votes": null
    },
    {
      "id": "2438763",
      "postDate": "09/14/2023 13:45:16",
      "content": "<ul>\n<li>1. Paper is very interesting and comes with code</li>\n<li>3. From third blog \"first, pre-train the model on a large quantity of unlabeled speech, then fine-tune on a smaller labeled dataset.\"  Very interesting!</li>\n<li>6. The analyticsvidhya website article is not readable unless one signs in.<br>\nThanks a lot for sharing.  </li>\n</ul>",
      "rawMarkdown": "1. Paper is very interesting and comes with code\n- 3. From third blog \"first, pre-train the model on a large quantity of unlabeled speech, then fine-tune on a smaller labeled dataset.\"  Very interesting!\n- 6. The analyticsvidhya website article is not readable unless one signs in.\nThanks a lot for sharing.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2438119,
      "author_name": "faysalmiah1721758",
      "author_url": "",
      "post_date": "09/14/2023 04:31:59",
      "content": "<p>Great resources <a href=\"https://www.kaggle.com/crsuthikshnkumar\" target=\"_blank\">@crsuthikshnkumar</a>, thanks for sharing.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2438763,
      "author_name": "robot2020",
      "author_url": "",
      "post_date": "09/14/2023 13:45:16",
      "content": "<ul>\n<li>1. Paper is very interesting and comes with code</li>\n<li>3. From third blog \"first, pre-train the model on a large quantity of unlabeled speech, then fine-tune on a smaller labeled dataset.\"  Very interesting!</li>\n<li>6. The analyticsvidhya website article is not readable unless one signs in.<br>\nThanks a lot for sharing.  </li>\n</ul>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2438100": "With respect to Bengali.AI Speech Recognition Competition, I observe the important model which \nis useful is Wav2Vec model.  I am sharing here the research papers which are relevant for the participants:\n\n1. WAV2VEC: UNSUPERVISED PRE-TRAINING FOR SPEECH RECOGNITION\nhttps://arxiv.org/pdf/1904.05862.pdf\n2.  Wav2Vec Model doc: https://huggingface.co/docs/transformers/model_doc/wav2vec2\n3.  An Illustrated Tour of Wav2Vec : https://jonathanbgn.com/2021/09/30/illustrated-wav2vec-2.html\n4. wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations\nhttps://proceedings.neurips.cc/paper/2020/file/92d1e1eb1cd6f9fba3227870bb6d7f07-Paper.pdf\n5. Speech to Text with Wav2Vec: https://www.kdnuggets.com/2021/03/speech-text-wav2vec.html\n6. Automatic Speech Recognition wth Wav2Vec: https://www.analyticsvidhya.com/blog/2022/06/automatic-speech-recognition-using-wav2vec2/",
    "2438119": "Great resources @crsuthikshnkumar, thanks for sharing.",
    "2438763": "1. Paper is very interesting and comes with code\n- 3. From third blog \"first, pre-train the model on a large quantity of unlabeled speech, then fine-tune on a smaller labeled dataset.\"  Very interesting!\n- 6. The analyticsvidhya website article is not readable unless one signs in.\nThanks a lot for sharing."
  },
  "source": "meta"
}