---
language:
- fy
base_model: distil-small.en
tags:
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_6_1
metrics:
- wer
model-index:
- name: DistilFT-Frisian-1h
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: mozilla-foundation/common_voice_6_fy_NL
      type: mozilla-foundation/common_voice_6_1
      args: 'config: fy-NL, split: train-1h'
    metrics:
    - name: Wer
      type: wer
      value: 77.86134378898592
---

<!-- 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. -->

# DistilFT-Frisian-1h

This model is a fine-tuned version of [distil-small.en](https://huggingface.co/distil-small.en) on the mozilla-foundation/common_voice_6_fy_NL dataset.
It achieves the following results on the evaluation set:
- Loss: 2.0756
- Wer: 77.8613

## 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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- training_steps: 2000
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch   | Step | Validation Loss | Wer     |
|:-------------:|:-------:|:----:|:---------------:|:-------:|
| 2.2753        | 5.6180  | 500  | 2.6854          | 87.9023 |
| 1.5355        | 11.2360 | 1000 | 2.2463          | 81.0515 |
| 1.201         | 16.8539 | 1500 | 2.1075          | 78.7239 |
| 1.0818        | 22.4719 | 2000 | 2.0756          | 77.8613 |


### Framework versions

- Transformers 4.41.0.dev0
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1