File size: 19,101 Bytes
2d1d0f8 377735e 2d1d0f8 377735e 2d1d0f8 377735e 2d1d0f8 377735e 2d1d0f8 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 |
---
language:
- he
license: apache-2.0
base_model: openai/whisper-small
tags:
- hf-asr-leaderboard
- generated_from_trainer
metrics:
- wer
model-index:
- name: he-cantillation
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# he-cantillation
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3963
- Wer: 16.4230
- Avg Precision Exact: 0.8621
- Avg Recall Exact: 0.8754
- Avg F1 Exact: 0.8679
- Avg Precision Letter Shift: 0.8906
- Avg Recall Letter Shift: 0.9042
- Avg F1 Letter Shift: 0.8966
- Avg Precision Word Level: 0.8954
- Avg Recall Word Level: 0.9066
- Avg F1 Word Level: 0.9002
- Avg Precision Word Shift: 0.9532
- Avg Recall Word Shift: 0.9639
- Avg F1 Word Shift: 0.9577
- Precision Median Exact: 0.9231
- Recall Median Exact: 0.9231
- F1 Median Exact: 0.9333
- 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: 0.6154
- Recall Min Word Shift: 0.7
- F1 Min Word Shift: 0.6667
## 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: 500
- training_steps: 200000
- 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 | 7.2500 | 103.0091 | 0.0 | 0.0 | 0.0 | 0.0002 | 0.0000 | 0.0000 | 0.0021 | 0.0165 | 0.0037 | 0.0285 | 0.0276 | 0.0273 | 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 |
| 0.0204 | 1.1457 | 10000 | 0.2495 | 20.8519 | 0.8158 | 0.8223 | 0.8182 | 0.8489 | 0.8558 | 0.8514 | 0.8532 | 0.8583 | 0.8549 | 0.9474 | 0.9544 | 0.9500 | 0.9091 | 0.9167 | 0.9091 | 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 | 0.6364 | 0.6364 |
| 0.0146 | 2.2915 | 20000 | 0.2638 | 19.4745 | 0.8330 | 0.8436 | 0.8375 | 0.8674 | 0.8788 | 0.8722 | 0.8717 | 0.8825 | 0.8763 | 0.9457 | 0.9559 | 0.9499 | 0.9091 | 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 | 0.4545 | 0.5 | 0.4762 |
| 0.0073 | 3.4372 | 30000 | 0.2853 | 18.3725 | 0.8473 | 0.8536 | 0.8497 | 0.8784 | 0.8855 | 0.8812 | 0.8825 | 0.8890 | 0.8850 | 0.9560 | 0.9615 | 0.9580 | 0.9167 | 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 | 0.5455 | 0.5455 | 0.5455 |
| 0.0041 | 4.5830 | 40000 | 0.3095 | 19.1354 | 0.8449 | 0.8556 | 0.8495 | 0.8788 | 0.8902 | 0.8836 | 0.8822 | 0.8931 | 0.8869 | 0.9487 | 0.9600 | 0.9535 | 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 | 0.4 | 0.4444 | 0.4211 |
| 0.0035 | 5.7287 | 50000 | 0.3130 | 18.7752 | 0.8527 | 0.8654 | 0.8582 | 0.8839 | 0.8979 | 0.8900 | 0.8889 | 0.9018 | 0.8945 | 0.9456 | 0.9593 | 0.9516 | 0.9167 | 0.9231 | 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 | 0.0 | 0.0 | 0.0 |
| 0.0031 | 6.8744 | 60000 | 0.3365 | 18.3513 | 0.8366 | 0.8516 | 0.8432 | 0.8664 | 0.8822 | 0.8734 | 0.8720 | 0.8859 | 0.8781 | 0.9446 | 0.9591 | 0.9509 | 0.9167 | 0.9231 | 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 | 0.0 | 0.0 | 0.0 |
| 0.0013 | 8.0202 | 70000 | 0.3247 | 17.4401 | 0.8558 | 0.8673 | 0.8607 | 0.8863 | 0.8981 | 0.8912 | 0.8912 | 0.9018 | 0.8956 | 0.9551 | 0.9680 | 0.9607 | 0.9167 | 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 | 0.7 | 0.75 | 0.7368 |
| 0.0012 | 9.1659 | 80000 | 0.3487 | 18.4361 | 0.8485 | 0.8602 | 0.8535 | 0.8847 | 0.8972 | 0.8901 | 0.8890 | 0.8994 | 0.8933 | 0.9528 | 0.9641 | 0.9576 | 0.9167 | 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 | 0.6154 | 0.7273 | 0.6667 |
| 0.0041 | 10.3116 | 90000 | 0.3509 | 17.4401 | 0.8443 | 0.8585 | 0.8506 | 0.8753 | 0.8900 | 0.8818 | 0.8805 | 0.8931 | 0.8860 | 0.9515 | 0.9659 | 0.9578 | 0.9167 | 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 | 0.5 | 0.5556 | 0.5263 |
| 0.0006 | 11.4574 | 100000 | 0.3491 | 17.2706 | 0.8562 | 0.8671 | 0.8608 | 0.8862 | 0.8985 | 0.8914 | 0.8905 | 0.9018 | 0.8953 | 0.9555 | 0.9678 | 0.9608 | 0.9167 | 0.9231 | 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 | 0.6923 | 0.7 | 0.7200 |
| 0.0001 | 12.6031 | 110000 | 0.3638 | 17.3342 | 0.8497 | 0.8605 | 0.8543 | 0.8779 | 0.8891 | 0.8826 | 0.8819 | 0.8914 | 0.8858 | 0.9556 | 0.9667 | 0.9602 | 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 | 0.6667 | 0.6667 | 0.6667 |
| 0.002 | 13.7489 | 120000 | 0.3561 | 17.2494 | 0.8465 | 0.8556 | 0.8503 | 0.8769 | 0.8865 | 0.8809 | 0.8808 | 0.8893 | 0.8843 | 0.9563 | 0.9659 | 0.9602 | 0.9167 | 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 | 0.7 | 0.7273 |
| 0.0012 | 14.8946 | 130000 | 0.3741 | 17.0587 | 0.8616 | 0.8756 | 0.8678 | 0.8932 | 0.9080 | 0.8998 | 0.8979 | 0.9111 | 0.9037 | 0.9561 | 0.9684 | 0.9615 | 0.9167 | 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 | 0.6923 | 0.6923 |
| 0.0015 | 16.0403 | 140000 | 0.3661 | 17.2070 | 0.8474 | 0.8603 | 0.8530 | 0.8775 | 0.8915 | 0.8837 | 0.8831 | 0.8940 | 0.8878 | 0.9521 | 0.9640 | 0.9572 | 0.9167 | 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 | 0.6 | 0.6667 | 0.6316 |
| 0.0001 | 17.1861 | 150000 | 0.3762 | 17.4825 | 0.8490 | 0.8600 | 0.8538 | 0.8803 | 0.8921 | 0.8854 | 0.8861 | 0.8957 | 0.8901 | 0.9542 | 0.9628 | 0.9577 | 0.9167 | 0.9231 | 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 | 0.6 | 0.6667 | 0.6316 |
| 0.0 | 18.3318 | 160000 | 0.3845 | 16.9951 | 0.8579 | 0.8683 | 0.8623 | 0.8874 | 0.8981 | 0.8919 | 0.8927 | 0.9013 | 0.8962 | 0.9525 | 0.9614 | 0.9561 | 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 | 0.6667 | 0.7 | 0.6957 |
| 0.0 | 19.4775 | 170000 | 0.3866 | 16.7620 | 0.8581 | 0.8717 | 0.8641 | 0.8889 | 0.9034 | 0.8953 | 0.8935 | 0.9058 | 0.8988 | 0.9551 | 0.9667 | 0.9601 | 0.9231 | 0.9231 | 0.9333 | 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 | 0.6667 | 0.6316 |
| 0.0 | 20.6233 | 180000 | 0.3849 | 16.5713 | 0.8577 | 0.8688 | 0.8624 | 0.8862 | 0.8977 | 0.8911 | 0.8917 | 0.9009 | 0.8955 | 0.9552 | 0.9643 | 0.9589 | 0.9231 | 0.9231 | 0.9412 | 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 | 0.7 | 0.6667 |
| 0.0 | 21.7690 | 190000 | 0.3930 | 16.5713 | 0.8611 | 0.8749 | 0.8671 | 0.8901 | 0.9044 | 0.8964 | 0.8959 | 0.9073 | 0.9007 | 0.9537 | 0.9644 | 0.9582 | 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 | 0.6154 | 0.7 | 0.6667 |
| 0.0 | 22.9148 | 200000 | 0.3963 | 16.4230 | 0.8621 | 0.8754 | 0.8679 | 0.8906 | 0.9042 | 0.8966 | 0.8954 | 0.9066 | 0.9002 | 0.9532 | 0.9639 | 0.9577 | 0.9231 | 0.9231 | 0.9333 | 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 | 0.7 | 0.6667 |
### Framework versions
- Transformers 4.41.2
- Pytorch 2.2.1
- Datasets 2.20.0
- Tokenizers 0.19.1
|