File size: 3,380 Bytes
3cc2f5b a2b277c 3cc2f5b a2b277c 3cc2f5b |
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 |
---
library_name: transformers
language:
- ar
license: apache-2.0
base_model: openai/whisper-base
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: Whisper base AR - BA
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. -->
# Whisper base AR - BA
This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the quran-ayat-speech-to-text dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0037
- Wer: 0.0550
- Cer: 0.0222
## 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: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- 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: 500
- num_epochs: 25
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|:-------------:|:-------:|:----:|:---------------:|:------:|:------:|
| 0.0049 | 1.0 | 282 | 0.0037 | 0.0498 | 0.0180 |
| 0.0031 | 2.0 | 564 | 0.0042 | 0.0525 | 0.0202 |
| 0.0015 | 3.0 | 846 | 0.0049 | 0.0568 | 0.0226 |
| 0.0012 | 4.0 | 1128 | 0.0058 | 0.0590 | 0.0229 |
| 0.0008 | 5.0 | 1410 | 0.0057 | 0.0635 | 0.0243 |
| 0.0007 | 6.0 | 1692 | 0.0063 | 0.0639 | 0.0223 |
| 0.0005 | 7.0 | 1974 | 0.0067 | 0.0610 | 0.0237 |
| 0.0005 | 8.0 | 2256 | 0.0070 | 0.0612 | 0.0232 |
| 0.0003 | 9.0 | 2538 | 0.0073 | 0.0626 | 0.0243 |
| 0.0003 | 10.0 | 2820 | 0.0080 | 0.0643 | 0.0239 |
| 0.0003 | 11.0 | 3102 | 0.0088 | 0.0635 | 0.0237 |
| 0.0003 | 12.0 | 3384 | 0.0087 | 0.0605 | 0.0231 |
| 0.0002 | 13.0 | 3666 | 0.0092 | 0.0612 | 0.0239 |
| 0.0002 | 14.0 | 3948 | 0.0104 | 0.0610 | 0.0226 |
| 0.0001 | 15.0 | 4230 | 0.0105 | 0.0543 | 0.0194 |
| 0.0001 | 16.0 | 4512 | 0.0111 | 0.0568 | 0.0216 |
| 0.0 | 17.0 | 4794 | 0.0124 | 0.0556 | 0.0206 |
| 0.0001 | 18.0 | 5076 | 0.0128 | 0.0539 | 0.0210 |
| 0.0 | 19.0 | 5358 | 0.0130 | 0.0530 | 0.0201 |
| 0.0 | 20.0 | 5640 | 0.0130 | 0.0539 | 0.0197 |
| 0.0 | 21.0 | 5922 | 0.0129 | 0.0536 | 0.0202 |
| 0.0 | 22.0 | 6204 | 0.0130 | 0.0541 | 0.0208 |
| 0.0 | 23.0 | 6486 | 0.0130 | 0.0547 | 0.0211 |
| 0.0 | 24.0 | 6768 | 0.0129 | 0.0605 | 0.0246 |
| 0.0 | 24.9130 | 7025 | 0.0130 | 0.0547 | 0.0210 |
### Framework versions
- Transformers 4.51.1
- Pytorch 2.5.1+cu124
- Datasets 3.5.0
- Tokenizers 0.21.0
|