---
library_name: transformers
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-small-CV_Fleurs_AMMI_ALFFA-sw-1hrs-v1
  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-CV_Fleurs_AMMI_ALFFA-sw-1hrs-v1

This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.0750
- Wer: 0.6608
- Cer: 0.3094

## 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: 4
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_HF with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 100
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    | Cer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
| 2.8725        | 1.0   | 72   | 1.4755          | 0.9300 | 0.3706 |
| 0.9122        | 2.0   | 144  | 1.2054          | 0.7693 | 0.3183 |
| 0.4127        | 3.0   | 216  | 1.1717          | 0.6751 | 0.2664 |
| 0.2043        | 4.0   | 288  | 1.2860          | 0.7312 | 0.3201 |
| 0.1703        | 5.0   | 360  | 1.3121          | 0.5544 | 0.2231 |
| 0.1509        | 6.0   | 432  | 1.3884          | 0.5587 | 0.2045 |
| 0.1646        | 7.0   | 504  | 1.5089          | 0.6426 | 0.2695 |
| 0.1761        | 8.0   | 576  | 1.5317          | 0.5963 | 0.2422 |
| 0.1725        | 9.0   | 648  | 1.5984          | 0.5842 | 0.2437 |
| 0.1968        | 10.0  | 720  | 1.7078          | 0.6239 | 0.2540 |
| 0.1949        | 11.0  | 792  | 1.8013          | 0.6542 | 0.2965 |
| 0.1738        | 12.0  | 864  | 1.8570          | 0.9835 | 0.6013 |
| 0.1511        | 13.0  | 936  | 1.9424          | 0.6105 | 0.2695 |
| 0.1478        | 14.0  | 1008 | 1.8473          | 0.7972 | 0.3916 |
| 0.127         | 15.0  | 1080 | 1.8845          | 0.7513 | 0.3836 |
| 0.1058        | 16.0  | 1152 | 2.0406          | 0.6778 | 0.3159 |
| 0.0982        | 17.0  | 1224 | 2.0612          | 0.6522 | 0.3179 |
| 0.0938        | 18.0  | 1296 | 2.0644          | 0.6445 | 0.3006 |
| 0.0832        | 19.0  | 1368 | 2.0671          | 0.8840 | 0.4900 |
| 0.0787        | 20.0  | 1440 | 2.0750          | 0.6608 | 0.3094 |


### Framework versions

- Transformers 4.48.0
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0