--- language: - he base_model: Teamim-large-v2_Random-True_date-06-06-2024_21-59-43 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 [Teamim-large-v2_Random-True_date-06-06-2024_21-59-43](https://huggingface.co/Teamim-large-v2_Random-True_date-06-06-2024_21-59-43) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.1326 - Wer: 13.4139 - Avg Precision Exact: 0.9006 - Avg Recall Exact: 0.9003 - Avg F1 Exact: 0.8998 - Avg Precision Letter Shift: 0.9238 - Avg Recall Letter Shift: 0.9240 - Avg F1 Letter Shift: 0.9232 - Avg Precision Word Level: 0.9256 - Avg Recall Word Level: 0.9251 - Avg F1 Word Level: 0.9246 - Avg Precision Word Shift: 0.9608 - Avg Recall Word Shift: 0.9621 - Avg F1 Word Shift: 0.9608 - Precision Median Exact: 0.9286 - Recall Median Exact: 0.9286 - F1 Median Exact: 0.9524 - 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: 0.6364 - Recall Min Word Shift: 0.5833 - F1 Min Word Shift: 0.6087 ## 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: 8 - eval_batch_size: 32 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 50 - training_steps: 8000 - 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 | 8e-05 | 1 | 1.4938 | 84.8485 | 0.2550 | 0.2358 | 0.2443 | 0.2950 | 0.2723 | 0.2823 | 0.3128 | 0.2852 | 0.2975 | 0.4888 | 0.4600 | 0.4720 | 0.2308 | 0.2143 | 0.2222 | 0.8333 | 0.8333 | 0.8333 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.2356 | 0.08 | 1000 | 0.1521 | 18.4573 | 0.8447 | 0.8494 | 0.8462 | 0.8707 | 0.8762 | 0.8727 | 0.8751 | 0.8788 | 0.8761 | 0.9351 | 0.9437 | 0.9385 | 0.9167 | 0.9167 | 0.9167 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.5385 | 0.5833 | 0.5600 | | 0.1781 | 0.16 | 2000 | 0.1429 | 16.1263 | 0.8883 | 0.8917 | 0.8892 | 0.9148 | 0.9188 | 0.9159 | 0.9165 | 0.9196 | 0.9173 | 0.9495 | 0.9556 | 0.9517 | 0.9231 | 0.9231 | 0.9231 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.6923 | 0.6667 | 0.6957 | | 0.1448 | 0.24 | 3000 | 0.1324 | 15.5118 | 0.8804 | 0.8821 | 0.8804 | 0.9043 | 0.9065 | 0.9046 | 0.9083 | 0.9095 | 0.9081 | 0.9537 | 0.9579 | 0.9550 | 0.9231 | 0.9231 | 0.9286 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.5 | 0.5556 | 0.5263 | | 0.1089 | 0.32 | 4000 | 0.1300 | 14.4522 | 0.8863 | 0.8888 | 0.8868 | 0.9069 | 0.9104 | 0.9079 | 0.9095 | 0.9134 | 0.9107 | 0.9530 | 0.9565 | 0.9539 | 0.9231 | 0.9231 | 0.9474 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.6154 | 0.5 | 0.5517 | | 0.0674 | 0.4 | 5000 | 0.1379 | 14.3463 | 0.8878 | 0.8942 | 0.8901 | 0.9122 | 0.9186 | 0.9146 | 0.9147 | 0.9203 | 0.9167 | 0.9563 | 0.9636 | 0.9591 | 0.9231 | 0.9286 | 0.9524 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.6364 | 0.5833 | 0.6087 | | 0.0548 | 0.48 | 6000 | 0.1380 | 13.8589 | 0.9005 | 0.9047 | 0.9018 | 0.9214 | 0.9262 | 0.9230 | 0.9239 | 0.9282 | 0.9253 | 0.9588 | 0.9650 | 0.9612 | 0.9286 | 0.9286 | 0.9524 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.6 | 0.5833 | 0.6087 | | 0.0461 | 0.56 | 7000 | 0.1343 | 13.6046 | 0.9012 | 0.9072 | 0.9034 | 0.9246 | 0.9309 | 0.9269 | 0.9268 | 0.9322 | 0.9288 | 0.9590 | 0.9646 | 0.9611 | 0.9286 | 0.9333 | 0.9524 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.6364 | 0.5833 | 0.6087 | | 0.0608 | 0.64 | 8000 | 0.1326 | 13.4139 | 0.9006 | 0.9003 | 0.8998 | 0.9238 | 0.9240 | 0.9232 | 0.9256 | 0.9251 | 0.9246 | 0.9608 | 0.9621 | 0.9608 | 0.9286 | 0.9286 | 0.9524 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.6364 | 0.5833 | 0.6087 | ### Framework versions - Transformers 4.42.0.dev0 - Pytorch 1.13.1+cu117 - Datasets 2.16.1 - Tokenizers 0.19.1