metadata
language:
- fa
license: apache-2.0
base_model: makhataei/Whisper-Small-Common-Voice
tags:
- fa-asr
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_16_0
metrics:
- wer
model-index:
- name: Whisper Small Persian
results: []
Whisper Small Persian
This model is a fine-tuned version of makhataei/Whisper-Small-Common-Voice on the Common Voice 16.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.8716
- Wer: 52.4832
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-08
- train_batch_size: 10
- eval_batch_size: 10
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 40
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- training_steps: 1000
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0002 | 0.16 | 100 | 0.8707 | 51.8989 |
| 0.0002 | 0.31 | 200 | 0.8713 | 52.1447 |
| 0.0002 | 0.47 | 300 | 0.8716 | 52.1377 |
| 0.0002 | 0.62 | 400 | 0.8717 | 52.2907 |
| 0.0015 | 0.78 | 500 | 0.8718 | 52.4762 |
| 0.0002 | 0.93 | 600 | 0.8716 | 52.4878 |
| 0.0002 | 1.09 | 700 | 0.8716 | 52.4901 |
| 0.0001 | 1.25 | 800 | 0.8716 | 52.4832 |
| 0.0002 | 1.4 | 900 | 0.8716 | 52.4832 |
| 0.0002 | 1.56 | 1000 | 0.8716 | 52.4832 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.0.1+cu117
- Datasets 2.15.0
- Tokenizers 0.15.0