--- library_name: transformers language: - he license: mit base_model: openai/whisper-large-v3-turbo tags: - hf-asr-leaderboard - generated_from_trainer metrics: - wer model-index: - name: he-cantillation results: [] --- # he-cantillation This model is a fine-tuned version of [openai/whisper-large-v3-turbo](https://huggingface.co/openai/whisper-large-v3-turbo) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.4313 - Wer: 37.8893 - Avg Precision Exact: 0.4754 - Avg Recall Exact: 0.4866 - Avg F1 Exact: 0.4799 - Avg Precision Letter Shift: 0.4984 - Avg Recall Letter Shift: 0.5113 - Avg F1 Letter Shift: 0.5030 - Avg Precision Word Level: 0.5105 - Avg Recall Word Level: 0.5244 - Avg F1 Word Level: 0.5151 - Avg Precision Word Shift: 0.6800 - Avg Recall Word Shift: 0.7089 - Avg F1 Word Shift: 0.6906 - Precision Median Exact: 0.375 - Recall Median Exact: 0.4068 - F1 Median Exact: 0.3905 - Precision Max Exact: 1.0 - Recall Max Exact: 1.0 - F1 Max Exact: 1.0 - Precision Min Exact: 0.0 - Recall Min Exact: 0.0 - F1 Min Exact: 0.0 - Precision Min Letter Shift: 0.0 - Recall Min Letter Shift: 0.0 - F1 Min Letter Shift: 0.0 - Precision Min Word Level: 0.0 - Recall Min Word Level: 0.0 - F1 Min Word Level: 0.0 - Precision Min Word Shift: 0.0 - Recall Min Word Shift: 0.0 - F1 Min Word Shift: 0.0 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 16 - eval_batch_size: 2 - seed: 42 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 1000 - training_steps: 60000 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | Avg Precision Exact | Avg Recall Exact | Avg F1 Exact | Avg Precision Letter Shift | Avg Recall Letter Shift | Avg F1 Letter Shift | Avg Precision Word Level | Avg Recall Word Level | Avg F1 Word Level | Avg Precision Word Shift | Avg Recall Word Shift | Avg F1 Word Shift | Precision Median Exact | Recall Median Exact | F1 Median Exact | Precision Max Exact | Recall Max Exact | F1 Max Exact | Precision Min Exact | Recall Min Exact | F1 Min Exact | Precision Min Letter Shift | Recall Min Letter Shift | F1 Min Letter Shift | Precision Min Word Level | Recall Min Word Level | F1 Min Word Level | Precision Min Word Shift | Recall Min Word Shift | F1 Min Word Shift | |:-------------:|:------:|:-----:|:---------------:|:--------:|:-------------------:|:----------------:|:------------:|:--------------------------:|:-----------------------:|:-------------------:|:------------------------:|:---------------------:|:-----------------:|:------------------------:|:---------------------:|:-----------------:|:----------------------:|:-------------------:|:---------------:|:-------------------:|:----------------:|:------------:|:-------------------:|:----------------:|:------------:|:--------------------------:|:-----------------------:|:-------------------:|:------------------------:|:---------------------:|:-----------------:|:------------------------:|:---------------------:|:-----------------:| | No log | 0.0001 | 1 | 6.2393 | 110.8685 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | | 0.1465 | 0.2962 | 2500 | 0.7718 | 61.2193 | 0.2875 | 0.3030 | 0.2936 | 0.3219 | 0.3423 | 0.3297 | 0.3405 | 0.3623 | 0.3483 | 0.5405 | 0.5932 | 0.5599 | 0.1368 | 0.1579 | 0.1455 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0667 | 0.5925 | 5000 | 0.6041 | 52.8951 | 0.3657 | 0.3756 | 0.3693 | 0.3989 | 0.4116 | 0.4035 | 0.4161 | 0.4290 | 0.4206 | 0.6173 | 0.6455 | 0.6280 | 0.2160 | 0.2333 | 0.2222 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0576 | 0.8887 | 7500 | 0.5474 | 49.1429 | 0.3890 | 0.3981 | 0.3923 | 0.4180 | 0.4302 | 0.4223 | 0.4327 | 0.4457 | 0.4370 | 0.6206 | 0.6476 | 0.6303 | 0.2261 | 0.2428 | 0.2353 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0552 | 1.1850 | 10000 | 0.5546 | 49.3049 | 0.3822 | 0.3988 | 0.3883 | 0.4092 | 0.4295 | 0.4165 | 0.4261 | 0.4496 | 0.4338 | 0.6053 | 0.6561 | 0.6226 | 0.2222 | 0.2576 | 0.2363 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0497 | 1.4812 | 12500 | 0.5479 | 49.9147 | 0.3636 | 0.3712 | 0.3666 | 0.3935 | 0.4041 | 0.3975 | 0.4113 | 0.4225 | 0.4152 | 0.6057 | 0.6373 | 0.6173 | 0.2 | 0.2094 | 0.2013 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0295 | 1.7775 | 15000 | 0.5342 | 48.7155 | 0.4055 | 0.4138 | 0.4083 | 0.4338 | 0.4439 | 0.4371 | 0.4500 | 0.4618 | 0.4535 | 0.6309 | 0.6577 | 0.6396 | 0.2470 | 0.2571 | 0.25 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0377 | 2.0737 | 17500 | 0.5136 | 46.1304 | 0.3919 | 0.4037 | 0.3967 | 0.4175 | 0.4322 | 0.4232 | 0.4320 | 0.4483 | 0.4378 | 0.6157 | 0.6509 | 0.6288 | 0.2099 | 0.2308 | 0.2175 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0202 | 2.3699 | 20000 | 0.5309 | 44.3608 | 0.4012 | 0.4122 | 0.4054 | 0.4276 | 0.4412 | 0.4325 | 0.4436 | 0.4596 | 0.4488 | 0.6299 | 0.6650 | 0.6427 | 0.2329 | 0.2576 | 0.2449 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.029 | 2.6662 | 22500 | 0.5077 | 46.8131 | 0.4103 | 0.4176 | 0.4129 | 0.4351 | 0.4440 | 0.4382 | 0.4502 | 0.4599 | 0.4530 | 0.6307 | 0.6557 | 0.6394 | 0.2275 | 0.25 | 0.2338 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0124 | 2.9624 | 25000 | 0.4679 | 43.1952 | 0.4293 | 0.4428 | 0.4344 | 0.4533 | 0.4699 | 0.4596 | 0.4672 | 0.4853 | 0.4735 | 0.6417 | 0.6792 | 0.6556 | 0.3 | 0.3333 | 0.3114 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0114 | 3.2587 | 27500 | 0.4747 | 47.7293 | 0.4276 | 0.4394 | 0.4321 | 0.4527 | 0.4666 | 0.4579 | 0.4666 | 0.4807 | 0.4714 | 0.6385 | 0.6666 | 0.6487 | 0.2812 | 0.3114 | 0.2963 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0265 | 3.5549 | 30000 | 0.4438 | 42.8144 | 0.4250 | 0.4368 | 0.4294 | 0.4479 | 0.4631 | 0.4535 | 0.4625 | 0.4799 | 0.4683 | 0.6470 | 0.6832 | 0.6599 | 0.2778 | 0.3012 | 0.2857 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0238 | 3.8512 | 32500 | 0.4454 | 41.6590 | 0.4508 | 0.4614 | 0.4549 | 0.4731 | 0.4859 | 0.4781 | 0.4868 | 0.5000 | 0.4919 | 0.6597 | 0.6896 | 0.6712 | 0.3191 | 0.3470 | 0.3276 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0178 | 4.1474 | 35000 | 0.4334 | 40.8610 | 0.4631 | 0.4709 | 0.4661 | 0.4880 | 0.4972 | 0.4912 | 0.5021 | 0.5124 | 0.5053 | 0.6759 | 0.6998 | 0.6842 | 0.3401 | 0.3581 | 0.3478 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0094 | 4.4437 | 37500 | 0.4594 | 42.0208 | 0.4477 | 0.4581 | 0.4516 | 0.4716 | 0.4853 | 0.4764 | 0.4855 | 0.5001 | 0.4903 | 0.6606 | 0.6936 | 0.6719 | 0.3125 | 0.3333 | 0.3182 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.008 | 4.7399 | 40000 | 0.4448 | 40.1024 | 0.4471 | 0.4582 | 0.4517 | 0.4702 | 0.4837 | 0.4754 | 0.4829 | 0.4978 | 0.4882 | 0.6558 | 0.6893 | 0.6685 | 0.3061 | 0.3333 | 0.3142 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0167 | 5.0361 | 42500 | 0.4367 | 42.5547 | 0.4564 | 0.4656 | 0.4600 | 0.4798 | 0.4902 | 0.4836 | 0.4923 | 0.5035 | 0.4964 | 0.6707 | 0.6940 | 0.6789 | 0.3333 | 0.3539 | 0.3422 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0028 | 5.3324 | 45000 | 0.4516 | 39.7654 | 0.4588 | 0.4687 | 0.4626 | 0.4807 | 0.4933 | 0.4855 | 0.4938 | 0.5078 | 0.4989 | 0.6760 | 0.7062 | 0.6872 | 0.3333 | 0.3600 | 0.3453 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0055 | 5.6286 | 47500 | 0.4354 | 38.9630 | 0.4715 | 0.4809 | 0.4752 | 0.4942 | 0.5057 | 0.4982 | 0.5061 | 0.5197 | 0.5104 | 0.6761 | 0.7034 | 0.6856 | 0.3556 | 0.3801 | 0.3636 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0196 | 5.9249 | 50000 | 0.4274 | 38.3824 | 0.4726 | 0.4855 | 0.4774 | 0.4945 | 0.5098 | 0.4999 | 0.5068 | 0.5234 | 0.5122 | 0.6692 | 0.7027 | 0.6811 | 0.375 | 0.4054 | 0.3823 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0038 | 6.2211 | 52500 | 0.4482 | 38.9543 | 0.4652 | 0.4763 | 0.4699 | 0.4880 | 0.5013 | 0.4935 | 0.5008 | 0.5155 | 0.5065 | 0.6727 | 0.7037 | 0.6846 | 0.3552 | 0.3846 | 0.3691 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0108 | 6.5174 | 55000 | 0.4453 | 39.4109 | 0.4742 | 0.4842 | 0.4783 | 0.4961 | 0.5079 | 0.5006 | 0.5083 | 0.5215 | 0.5128 | 0.6771 | 0.7048 | 0.6869 | 0.3624 | 0.3939 | 0.3787 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0078 | 6.8136 | 57500 | 0.4373 | 38.4320 | 0.4784 | 0.4893 | 0.4828 | 0.5000 | 0.5133 | 0.5048 | 0.5132 | 0.5276 | 0.5180 | 0.6806 | 0.7116 | 0.6918 | 0.3797 | 0.4138 | 0.4000 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0124 | 7.1098 | 60000 | 0.4313 | 37.8893 | 0.4754 | 0.4866 | 0.4799 | 0.4984 | 0.5113 | 0.5030 | 0.5105 | 0.5244 | 0.5151 | 0.6800 | 0.7089 | 0.6906 | 0.375 | 0.4068 | 0.3905 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | ### Framework versions - Transformers 4.49.0 - Pytorch 2.6.0+cu126 - Datasets 2.12.0 - Tokenizers 0.20.1