---
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: 0.7811
- Wer: 43.9230

## 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-06
- 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     |
|:-------------:|:-----:|:----:|:---------------:|:-------:|
| 0.0014        | 0.16  | 100  | 0.7112          | 43.9740 |
| 0.0048        | 0.31  | 200  | 0.7077          | 42.5968 |
| 0.0048        | 0.47  | 300  | 0.7099          | 42.8982 |
| 0.0032        | 0.62  | 400  | 0.7116          | 43.4802 |
| 0.0041        | 0.78  | 500  | 0.7156          | 43.6935 |
| 0.0025        | 0.93  | 600  | 0.7184          | 43.5660 |
| 0.0012        | 1.09  | 700  | 0.7212          | 44.5421 |
| 0.0005        | 1.25  | 800  | 0.7338          | 43.8164 |
| 0.0014        | 1.4   | 900  | 0.7409          | 44.0529 |
| 0.0009        | 1.56  | 1000 | 0.7414          | 44.3010 |
| 0.001         | 1.71  | 1100 | 0.7408          | 45.0475 |
| 0.0006        | 1.87  | 1200 | 0.7447          | 44.6812 |
| 0.0005        | 2.03  | 1300 | 0.7493          | 43.4268 |
| 0.0005        | 2.18  | 1400 | 0.7564          | 43.2877 |
| 0.0004        | 2.34  | 1500 | 0.7580          | 43.8280 |
| 0.0004        | 2.49  | 1600 | 0.7626          | 43.5173 |
| 0.019         | 2.65  | 1700 | 0.7654          | 43.7677 |
| 0.0004        | 2.8   | 1800 | 0.7679          | 43.8001 |
| 0.0003        | 2.96  | 1900 | 0.7694          | 43.7491 |
| 0.0003        | 3.12  | 2000 | 0.7713          | 43.8906 |
| 0.0003        | 3.27  | 2100 | 0.7737          | 43.7978 |
| 0.0003        | 3.43  | 2200 | 0.7763          | 43.6031 |
| 0.0003        | 3.58  | 2300 | 0.7770          | 43.9253 |
| 0.0003        | 3.74  | 2400 | 0.7778          | 43.9995 |
| 0.0003        | 3.9   | 2500 | 0.7780          | 44.0946 |
| 0.0003        | 4.05  | 2600 | 0.7787          | 43.8906 |
| 0.0156        | 4.21  | 2700 | 0.7798          | 43.9091 |
| 0.0003        | 4.36  | 2800 | 0.7805          | 43.9114 |
| 0.0003        | 4.52  | 2900 | 0.7809          | 43.9230 |
| 0.0003        | 4.67  | 3000 | 0.7811          | 43.9230 |


### Framework versions

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