File size: 11,696 Bytes
24b9c8f a390a99 24b9c8f a390a99 24b9c8f a390a99 24b9c8f a390a99 24b9c8f |
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 |
---
language:
- he
license: apache-2.0
base_model: openai/whisper-tiny
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-tiny](https://huggingface.co/openai/whisper-tiny) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2552
- Wer: 16.1604
- Avg Precision Exact: 0.8115
- Avg Recall Exact: 0.8113
- Avg F1 Exact: 0.8110
- Avg Precision Letter Shift: 0.8369
- Avg Recall Letter Shift: 0.8369
- Avg F1 Letter Shift: 0.8365
- Avg Precision Word Level: 0.8414
- Avg Recall Word Level: 0.8413
- Avg F1 Word Level: 0.8409
- Avg Precision Word Shift: 0.9648
- Avg Recall Word Shift: 0.9659
- Avg F1 Word Shift: 0.9648
- Precision Median Exact: 0.9286
- Recall Median Exact: 0.9286
- F1 Median Exact: 0.9286
- 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: 0.0001
- 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: 100000
- 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 | 7.9394 | 100.8130 | 0.0002 | 0.0009 | 0.0004 | 0.0043 | 0.0034 | 0.0034 | 0.0036 | 0.0204 | 0.0059 | 0.0398 | 0.0360 | 0.0353 | 0.0 | 0.0 | 0.0 | 0.1111 | 0.5 | 0.1538 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
| 0.0573 | 0.8 | 10000 | 0.1857 | 22.9823 | 0.7811 | 0.7770 | 0.7785 | 0.8163 | 0.8120 | 0.8135 | 0.8236 | 0.8193 | 0.8208 | 0.9437 | 0.9406 | 0.9414 | 0.9 | 0.8889 | 0.8889 | 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 |
| 0.0276 | 1.6 | 20000 | 0.2057 | 19.9224 | 0.8052 | 0.8105 | 0.8073 | 0.8336 | 0.8391 | 0.8357 | 0.8390 | 0.8441 | 0.8410 | 0.9451 | 0.9521 | 0.9478 | 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 | 0.0 | 0.0 | 0.0 |
| 0.0238 | 2.4 | 30000 | 0.2099 | 19.0096 | 0.8151 | 0.8152 | 0.8147 | 0.8437 | 0.8440 | 0.8434 | 0.8491 | 0.8495 | 0.8488 | 0.9557 | 0.9579 | 0.9562 | 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 | 0.1429 | 0.125 | 0.1333 |
| 0.0157 | 3.2 | 40000 | 0.2171 | 18.0414 | 0.8332 | 0.8294 | 0.8309 | 0.8616 | 0.8580 | 0.8593 | 0.8671 | 0.8641 | 0.8651 | 0.9607 | 0.9587 | 0.9591 | 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 | 0.0769 | 0.0714 | 0.0741 |
| 0.0061 | 4.0 | 50000 | 0.2365 | 17.8123 | 0.8326 | 0.8322 | 0.8320 | 0.8597 | 0.8592 | 0.8590 | 0.8658 | 0.8651 | 0.8650 | 0.9598 | 0.9604 | 0.9595 | 0.9286 | 0.9286 | 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 | 0.0833 | 0.0769 | 0.08 |
| 0.0038 | 4.8 | 60000 | 0.2350 | 17.3651 | 0.8243 | 0.8251 | 0.8242 | 0.8520 | 0.8529 | 0.8520 | 0.8568 | 0.8578 | 0.8568 | 0.9571 | 0.9600 | 0.9579 | 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 | 0.0 | 0.0 | 0.0 |
| 0.002 | 5.6 | 70000 | 0.2404 | 17.0288 | 0.8319 | 0.8323 | 0.8317 | 0.8592 | 0.8595 | 0.8589 | 0.8649 | 0.8652 | 0.8645 | 0.9609 | 0.9620 | 0.9608 | 0.9286 | 0.9286 | 0.9310 | 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 |
| 0.002 | 6.4 | 80000 | 0.2439 | 17.0251 | 0.8116 | 0.8127 | 0.8117 | 0.8368 | 0.8379 | 0.8369 | 0.8420 | 0.8429 | 0.8420 | 0.9567 | 0.9592 | 0.9573 | 0.9231 | 0.9258 | 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 | 0.0 | 0.0 | 0.0 |
| 0.001 | 7.2 | 90000 | 0.2548 | 16.8995 | 0.8106 | 0.8102 | 0.8100 | 0.8368 | 0.8364 | 0.8362 | 0.8421 | 0.8414 | 0.8413 | 0.9575 | 0.9579 | 0.9571 | 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 | 0.0 | 0.0 | 0.0 |
| 0.0001 | 8.0 | 100000 | 0.2552 | 16.1604 | 0.8115 | 0.8113 | 0.8110 | 0.8369 | 0.8369 | 0.8365 | 0.8414 | 0.8413 | 0.8409 | 0.9648 | 0.9659 | 0.9648 | 0.9286 | 0.9286 | 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 | 0.0 | 0.0 | 0.0 |
### Framework versions
- Transformers 4.41.2
- Pytorch 2.2.1
- Datasets 2.20.0
- Tokenizers 0.19.1
|