Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,89 @@
|
|
| 1 |
-
---
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
language:
|
| 4 |
+
- ar
|
| 5 |
+
license: apache-2.0
|
| 6 |
+
base_model: openai/whisper-small
|
| 7 |
+
tags:
|
| 8 |
+
- whisper-event
|
| 9 |
+
- generated_from_trainer
|
| 10 |
+
datasets:
|
| 11 |
+
- mozilla-foundation/common_voice_11_0
|
| 12 |
+
metrics:
|
| 13 |
+
- wer
|
| 14 |
+
model-index:
|
| 15 |
+
- name: Whisper Small AR - Mohammed Bakheet
|
| 16 |
+
results:
|
| 17 |
+
- task:
|
| 18 |
+
name: Automatic Speech Recognition
|
| 19 |
+
type: automatic-speech-recognition
|
| 20 |
+
dataset:
|
| 21 |
+
name: Common Voice 11.0
|
| 22 |
+
type: mozilla-foundation/common_voice_11_0
|
| 23 |
+
config: ar
|
| 24 |
+
split: test
|
| 25 |
+
args: ar
|
| 26 |
+
metrics:
|
| 27 |
+
- name: Wer
|
| 28 |
+
type: wer
|
| 29 |
+
value: 20.45616669795382
|
| 30 |
+
---
|
| 31 |
+
|
| 32 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 33 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 34 |
+
|
| 35 |
+
# Whisper Small AR - Mohammed Bakheet
|
| 36 |
+
|
| 37 |
+
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 11.0 dataset.
|
| 38 |
+
It achieves the following results on the evaluation set:
|
| 39 |
+
- Loss: 0.2601
|
| 40 |
+
- Wer: 20.4562
|
| 41 |
+
|
| 42 |
+
## Model description
|
| 43 |
+
|
| 44 |
+
More information needed
|
| 45 |
+
|
| 46 |
+
## Intended uses & limitations
|
| 47 |
+
|
| 48 |
+
More information needed
|
| 49 |
+
|
| 50 |
+
## Training and evaluation data
|
| 51 |
+
|
| 52 |
+
More information needed
|
| 53 |
+
|
| 54 |
+
## Training procedure
|
| 55 |
+
|
| 56 |
+
### Training hyperparameters
|
| 57 |
+
|
| 58 |
+
The following hyperparameters were used during training:
|
| 59 |
+
- learning_rate: 1e-05
|
| 60 |
+
- train_batch_size: 2
|
| 61 |
+
- eval_batch_size: 8
|
| 62 |
+
- seed: 42
|
| 63 |
+
- gradient_accumulation_steps: 16
|
| 64 |
+
- total_train_batch_size: 32
|
| 65 |
+
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 66 |
+
- lr_scheduler_type: linear
|
| 67 |
+
- lr_scheduler_warmup_steps: 500
|
| 68 |
+
- training_steps: 5000
|
| 69 |
+
- mixed_precision_training: Native AMP
|
| 70 |
+
|
| 71 |
+
### Training results
|
| 72 |
+
|
| 73 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 74 |
+
|:-------------:|:------:|:----:|:---------------:|:-------:|
|
| 75 |
+
| 0.5279 | 0.4158 | 500 | 0.3311 | 27.6591 |
|
| 76 |
+
| 0.2513 | 0.8316 | 1000 | 0.2866 | 24.5504 |
|
| 77 |
+
| 0.1673 | 1.2478 | 1500 | 0.2735 | 22.8928 |
|
| 78 |
+
| 0.1324 | 1.6635 | 2000 | 0.2645 | 21.8153 |
|
| 79 |
+
| 0.1138 | 2.0797 | 2500 | 0.2613 | 21.3816 |
|
| 80 |
+
| 0.064 | 2.4955 | 3000 | 0.2651 | 21.0006 |
|
| 81 |
+
| 0.0615 | 2.9113 | 3500 | 0.2601 | 20.4562 |
|
| 82 |
+
|
| 83 |
+
|
| 84 |
+
### Framework versions
|
| 85 |
+
|
| 86 |
+
- Transformers 4.46.2
|
| 87 |
+
- Pytorch 2.5.1+cu124
|
| 88 |
+
- Datasets 3.1.0
|
| 89 |
+
- Tokenizers 0.20.3
|