cantillation's picture
End of training
e6a7005 verified
---
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: 0.2146
- Wer: 12.3134
- Avg Precision Exact: 0.8990
- Avg Recall Exact: 0.9012
- Avg F1 Exact: 0.8997
- Avg Precision Letter Shift: 0.9196
- Avg Recall Letter Shift: 0.9219
- Avg F1 Letter Shift: 0.9204
- Avg Precision Word Level: 0.9216
- Avg Recall Word Level: 0.9240
- Avg F1 Word Level: 0.9224
- Avg Precision Word Shift: 0.9730
- Avg Recall Word Shift: 0.9761
- Avg F1 Word Shift: 0.9741
- Precision Median Exact: 1.0
- Recall Median Exact: 1.0
- F1 Median Exact: 1.0
- 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: 0.1429
- Recall Min Word Shift: 0.1
- F1 Min Word Shift: 0.1176
## 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 | 8e-05 | 1 | 7.1551 | 102.7088 | 0.0004 | 0.0034 | 0.0008 | 0.0169 | 0.0167 | 0.0165 | 0.0045 | 0.0485 | 0.0082 | 0.0865 | 0.0873 | 0.0863 | 0.0 | 0.0 | 0.0 | 0.1111 | 1.0 | 0.2000 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
| 0.0271 | 0.8 | 10000 | 0.1443 | 15.7539 | 0.8698 | 0.8704 | 0.8696 | 0.8921 | 0.8928 | 0.8920 | 0.8951 | 0.8962 | 0.8951 | 0.9611 | 0.9641 | 0.9620 | 0.9333 | 0.9333 | 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 | 0.1429 | 0.125 | 0.1333 |
| 0.0138 | 1.6 | 20000 | 0.1523 | 14.3570 | 0.8784 | 0.8793 | 0.8784 | 0.9010 | 0.9020 | 0.9010 | 0.9042 | 0.9053 | 0.9043 | 0.9672 | 0.9691 | 0.9676 | 0.9737 | 1.0 | 0.9565 | 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 | 0.0769 | 0.0769 |
| 0.0059 | 2.4 | 30000 | 0.1584 | 17.4242 | 0.8838 | 0.8843 | 0.8835 | 0.9054 | 0.9061 | 0.9051 | 0.9077 | 0.9090 | 0.9076 | 0.9658 | 0.9681 | 0.9661 | 1.0 | 1.0 | 0.9655 | 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 | 3.2 | 40000 | 0.1696 | 13.3925 | 0.8917 | 0.8933 | 0.8921 | 0.9129 | 0.9147 | 0.9134 | 0.9155 | 0.9171 | 0.9159 | 0.9724 | 0.9743 | 0.9729 | 1.0 | 1.0 | 0.9677 | 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 | 0.1 | 0.1176 |
| 0.0039 | 4.0 | 50000 | 0.1695 | 13.2779 | 0.8819 | 0.8816 | 0.8814 | 0.9029 | 0.9027 | 0.9024 | 0.9052 | 0.9052 | 0.9048 | 0.9707 | 0.9714 | 0.9706 | 1.0 | 1.0 | 0.9697 | 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 |
| 0.0027 | 4.8 | 60000 | 0.1823 | 13.4294 | 0.8837 | 0.8852 | 0.8841 | 0.9047 | 0.9066 | 0.9052 | 0.9073 | 0.9094 | 0.9078 | 0.9699 | 0.9729 | 0.9708 | 1.0 | 1.0 | 1.0 | 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 | 0.1 | 0.1176 |
| 0.0008 | 5.6 | 70000 | 0.1762 | 13.0340 | 0.8870 | 0.8902 | 0.8883 | 0.9071 | 0.9104 | 0.9084 | 0.9097 | 0.9131 | 0.9110 | 0.9710 | 0.9751 | 0.9726 | 1.0 | 1.0 | 1.0 | 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 | 0.1111 | 0.125 |
| 0.0006 | 6.4 | 80000 | 0.1845 | 12.8012 | 0.8926 | 0.8943 | 0.8931 | 0.9131 | 0.9150 | 0.9137 | 0.9155 | 0.9174 | 0.9161 | 0.9729 | 0.9753 | 0.9737 | 1.0 | 1.0 | 1.0 | 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 | 0.1 | 0.1176 |
| 0.0009 | 7.2 | 90000 | 0.1883 | 12.6497 | 0.8960 | 0.8943 | 0.8948 | 0.9173 | 0.9158 | 0.9162 | 0.9197 | 0.9184 | 0.9187 | 0.9745 | 0.9747 | 0.9742 | 1.0 | 1.0 | 1.0 | 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 | 0.1 | 0.1176 |
| 0.0003 | 8.0 | 100000 | 0.1888 | 12.3688 | 0.8960 | 0.8977 | 0.8965 | 0.9169 | 0.9186 | 0.9174 | 0.9192 | 0.9211 | 0.9198 | 0.9745 | 0.9772 | 0.9755 | 1.0 | 1.0 | 1.0 | 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 |
| 0.0023 | 8.8 | 110000 | 0.1934 | 12.5240 | 0.8962 | 0.8948 | 0.8951 | 0.9162 | 0.9148 | 0.9151 | 0.9184 | 0.9172 | 0.9174 | 0.9737 | 0.9740 | 0.9734 | 1.0 | 1.0 | 1.0 | 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 | 0.1 | 0.1176 |
| 0.0005 | 9.6 | 120000 | 0.2024 | 12.6534 | 0.8919 | 0.8935 | 0.8924 | 0.9122 | 0.9139 | 0.9127 | 0.9146 | 0.9163 | 0.9151 | 0.9720 | 0.9745 | 0.9728 | 1.0 | 1.0 | 1.0 | 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 | 0.1111 | 0.125 |
| 0.0001 | 10.4 | 130000 | 0.1999 | 13.0229 | 0.8901 | 0.8925 | 0.8909 | 0.9109 | 0.9135 | 0.9118 | 0.9134 | 0.9160 | 0.9143 | 0.9714 | 0.9749 | 0.9727 | 1.0 | 1.0 | 1.0 | 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 |
| 0.0002 | 11.2 | 140000 | 0.2010 | 12.9675 | 0.8873 | 0.8920 | 0.8892 | 0.9085 | 0.9135 | 0.9106 | 0.9112 | 0.9163 | 0.9133 | 0.9698 | 0.9752 | 0.9720 | 1.0 | 1.0 | 1.0 | 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 |
| 0.0001 | 12.0 | 150000 | 0.2078 | 12.8455 | 0.8932 | 0.8961 | 0.8943 | 0.9149 | 0.9181 | 0.9161 | 0.9170 | 0.9202 | 0.9182 | 0.9724 | 0.9762 | 0.9739 | 1.0 | 1.0 | 1.0 | 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 | 0.1111 | 0.125 |
| 0.0001 | 12.8 | 160000 | 0.2063 | 12.6718 | 0.8950 | 0.8962 | 0.8952 | 0.9161 | 0.9175 | 0.9164 | 0.9183 | 0.9198 | 0.9187 | 0.9731 | 0.9757 | 0.9740 | 1.0 | 1.0 | 1.0 | 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 | 0.1111 | 0.125 |
| 0.0005 | 13.6 | 170000 | 0.2137 | 12.5203 | 0.8941 | 0.8960 | 0.8947 | 0.9143 | 0.9164 | 0.9150 | 0.9165 | 0.9186 | 0.9172 | 0.9712 | 0.9742 | 0.9723 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0909 | 0.1 | 0.1000 |
| 0.0002 | 14.4 | 180000 | 0.2121 | 12.5979 | 0.8945 | 0.8964 | 0.8951 | 0.9155 | 0.9176 | 0.9162 | 0.9178 | 0.9199 | 0.9185 | 0.9731 | 0.9760 | 0.9741 | 1.0 | 1.0 | 1.0 | 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 | 0.1111 | 0.125 |
| 0.0 | 15.2 | 190000 | 0.2143 | 12.3282 | 0.8985 | 0.9008 | 0.8993 | 0.9193 | 0.9218 | 0.9202 | 0.9215 | 0.9240 | 0.9224 | 0.9734 | 0.9766 | 0.9746 | 1.0 | 1.0 | 1.0 | 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 |
| 0.0 | 16.0 | 200000 | 0.2146 | 12.3134 | 0.8990 | 0.9012 | 0.8997 | 0.9196 | 0.9219 | 0.9204 | 0.9216 | 0.9240 | 0.9224 | 0.9730 | 0.9761 | 0.9741 | 1.0 | 1.0 | 1.0 | 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 | 0.1 | 0.1176 |
### Framework versions
- Transformers 4.41.2
- Pytorch 2.2.1
- Datasets 2.20.0
- Tokenizers 0.19.1