{
  "id": 491272,
  "title": "Existing Bengali ASR Models",
  "url": "/competitions/ben10/discussion/491272",
  "author_name": "Tahsin",
  "post_date": "2024-04-05T09:21:04.726000",
  "votes": 7,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Let's dump the existing ASR models capable of transcribing bengali speech in this post.</p>\n<p>Models from the Bengali.AI ASR comp from last year</p>\n<p>1st (base model: Whisper), <a href=\"https://www.kaggle.com/competitions/bengaliai-speech/discussion/447961\" target=\"_blank\">https://www.kaggle.com/competitions/bengaliai-speech/discussion/447961</a><br>\n2nd (base model: ai4bharat/IndicBERTv2-MLM-Sam-TLM), <a href=\"https://www.kaggle.com/competitions/bengaliai-speech/discussion/447976\" target=\"_blank\">https://www.kaggle.com/competitions/bengaliai-speech/discussion/447976</a><br>\n3rd (base model: ai4bharat/indicwav2vec_v1_bengali), <a href=\"https://www.kaggle.com/competitions/bengaliai-speech/discussion/447957\" target=\"_blank\">https://www.kaggle.com/competitions/bengaliai-speech/discussion/447957</a><br>\n5th (base model: ensemble of IndicWav2Vec and XLS-R), <a href=\"https://www.kaggle.com/competitions/bengaliai-speech/discussion/448006\" target=\"_blank\">https://www.kaggle.com/competitions/bengaliai-speech/discussion/448006</a></p>\n<p>YellowKing model, <a href=\"https://www.kaggle.com/competitions/bengaliai-speech/discussion/425457\" target=\"_blank\">https://www.kaggle.com/competitions/bengaliai-speech/discussion/425457</a></p>\n<p>BanglaConformer, <a href=\"https://huggingface.co/bengaliAI/BanglaConformer\" target=\"_blank\">https://huggingface.co/bengaliAI/BanglaConformer</a></p>\n<p>Hishab BN FastConformer, <a href=\"https://huggingface.co/hishab/hishab_bn_fastconformer\" target=\"_blank\">https://huggingface.co/hishab/hishab_bn_fastconformer</a></p>",
  "messages": [
    {
      "id": 2736574,
      "postDate": "2024-04-05T09:21:04.727Z",
      "content": "<p>Let's dump the existing ASR models capable of transcribing bengali speech in this post.</p>\n<p>Models from the Bengali.AI ASR comp from last year</p>\n<p>1st (base model: Whisper), <a href=\"https://www.kaggle.com/competitions/bengaliai-speech/discussion/447961\" target=\"_blank\">https://www.kaggle.com/competitions/bengaliai-speech/discussion/447961</a><br>\n2nd (base model: ai4bharat/IndicBERTv2-MLM-Sam-TLM), <a href=\"https://www.kaggle.com/competitions/bengaliai-speech/discussion/447976\" target=\"_blank\">https://www.kaggle.com/competitions/bengaliai-speech/discussion/447976</a><br>\n3rd (base model: ai4bharat/indicwav2vec_v1_bengali), <a href=\"https://www.kaggle.com/competitions/bengaliai-speech/discussion/447957\" target=\"_blank\">https://www.kaggle.com/competitions/bengaliai-speech/discussion/447957</a><br>\n5th (base model: ensemble of IndicWav2Vec and XLS-R), <a href=\"https://www.kaggle.com/competitions/bengaliai-speech/discussion/448006\" target=\"_blank\">https://www.kaggle.com/competitions/bengaliai-speech/discussion/448006</a></p>\n<p>YellowKing model, <a href=\"https://www.kaggle.com/competitions/bengaliai-speech/discussion/425457\" target=\"_blank\">https://www.kaggle.com/competitions/bengaliai-speech/discussion/425457</a></p>\n<p>BanglaConformer, <a href=\"https://huggingface.co/bengaliAI/BanglaConformer\" target=\"_blank\">https://huggingface.co/bengaliAI/BanglaConformer</a></p>\n<p>Hishab BN FastConformer, <a href=\"https://huggingface.co/hishab/hishab_bn_fastconformer\" target=\"_blank\">https://huggingface.co/hishab/hishab_bn_fastconformer</a></p>",
      "rawMarkdown": "Let's dump the existing ASR models capable of transcribing bengali speech in this post.\n\nModels from the Bengali.AI ASR comp from last year\n\n1st (base model: Whisper), https://www.kaggle.com/competitions/bengaliai-speech/discussion/447961\n2nd (base model: ai4bharat/IndicBERTv2-MLM-Sam-TLM), https://www.kaggle.com/competitions/bengaliai-speech/discussion/447976\n3rd (base model: ai4bharat/indicwav2vec_v1_bengali), https://www.kaggle.com/competitions/bengaliai-speech/discussion/447957\n5th (base model: ensemble of IndicWav2Vec and XLS-R), https://www.kaggle.com/competitions/bengaliai-speech/discussion/448006\n\nYellowKing model, https://www.kaggle.com/competitions/bengaliai-speech/discussion/425457\n\nBanglaConformer, https://huggingface.co/bengaliAI/BanglaConformer\n\nHishab BN FastConformer, https://huggingface.co/hishab/hishab_bn_fastconformer\n\n",
      "votes": 7
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "2736574": "Let's dump the existing ASR models capable of transcribing bengali speech in this post.\n\nModels from the Bengali.AI ASR comp from last year\n\n1st (base model: Whisper), https://www.kaggle.com/competitions/bengaliai-speech/discussion/447961\n2nd (base model: ai4bharat/IndicBERTv2-MLM-Sam-TLM), https://www.kaggle.com/competitions/bengaliai-speech/discussion/447976\n3rd (base model: ai4bharat/indicwav2vec_v1_bengali), https://www.kaggle.com/competitions/bengaliai-speech/discussion/447957\n5th (base model: ensemble of IndicWav2Vec and XLS-R), https://www.kaggle.com/competitions/bengaliai-speech/discussion/448006\n\nYellowKing model, https://www.kaggle.com/competitions/bengaliai-speech/discussion/425457\n\nBanglaConformer, https://huggingface.co/bengaliAI/BanglaConformer\n\nHishab BN FastConformer, https://huggingface.co/hishab/hishab_bn_fastconformer\n\n"
  }
}