---
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: []
---

<!-- 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 Small Persian

This model is a fine-tuned version of [makhataei/Whisper-Small-Common-Voice](https://huggingface.co/makhataei/Whisper-Small-Common-Voice) on the Common Voice 16.0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7773
- Wer: 58.2147

## 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: 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: 3000

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer     |
|:-------------:|:-----:|:----:|:---------------:|:-------:|
| 0.0106        | 0.16  | 100  | 0.7024          | 44.3125 |
| 0.0107        | 0.31  | 200  | 0.7246          | 46.4665 |
| 0.0121        | 0.47  | 300  | 0.7213          | 44.8249 |
| 0.0096        | 0.62  | 400  | 0.7252          | 48.7294 |
| 0.0118        | 0.78  | 500  | 0.7118          | 47.6837 |
| 0.011         | 0.93  | 600  | 0.7221          | 47.6119 |
| 0.0058        | 1.09  | 700  | 0.7494          | 46.9117 |
| 0.0051        | 1.25  | 800  | 0.7611          | 46.6844 |
| 0.009         | 1.4   | 900  | 0.7271          | 47.2919 |
| 0.0077        | 1.56  | 1000 | 0.7287          | 58.6274 |
| 0.0058        | 1.71  | 1100 | 0.7582          | 64.5653 |
| 0.0067        | 1.87  | 1200 | 0.7135          | 66.7841 |
| 0.0031        | 2.03  | 1300 | 0.7395          | 61.1755 |
| 0.0033        | 2.18  | 1400 | 0.7353          | 62.7568 |
| 0.0025        | 2.34  | 1500 | 0.7535          | 68.1080 |
| 0.0022        | 2.49  | 1600 | 0.7521          | 63.5126 |
| 0.0151        | 2.65  | 1700 | 0.7546          | 71.5140 |
| 0.0022        | 2.8   | 1800 | 0.7470          | 79.6058 |
| 0.0017        | 2.96  | 1900 | 0.7286          | 70.0348 |
| 0.0008        | 3.12  | 2000 | 0.7544          | 66.4433 |
| 0.0007        | 3.27  | 2100 | 0.7618          | 60.4336 |
| 0.0004        | 3.43  | 2200 | 0.7672          | 64.8505 |
| 0.0015        | 3.58  | 2300 | 0.7853          | 60.2017 |
| 0.0008        | 3.74  | 2400 | 0.7719          | 59.2233 |
| 0.0003        | 3.9   | 2500 | 0.7537          | 60.1414 |
| 0.0002        | 4.05  | 2600 | 0.7656          | 59.9490 |
| 0.0087        | 4.21  | 2700 | 0.7726          | 59.2140 |
| 0.0003        | 4.36  | 2800 | 0.7750          | 59.6661 |
| 0.0002        | 4.52  | 2900 | 0.7766          | 58.8036 |
| 0.0002        | 4.67  | 3000 | 0.7773          | 58.2147 |


### Framework versions

- Transformers 4.35.2
- Pytorch 2.0.1+cu117
- Datasets 2.15.0
- Tokenizers 0.15.0