---
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- google/fleurs
metrics:
- wer
model-index:
- name: xls-r-fleurs_nl-run2
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: audiofolder
      type: audiofolder
      config: default
      split: validation
      args: default
    metrics:
    - name: Wer
      type: wer
      value: 0.38678223185265437
---

<!-- 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. -->

# xls-r-fleurs_nl-run2

This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the FLEURS (nl) dataset.
It achieves the following results:
- Wer (Validation): 38.05%
- Wer (Test): 37.20%

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 30
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 7.4175        | 0.55  | 100  | 3.8140          | 1.0    |
| 3.2001        | 1.1   | 200  | 2.9283          | 1.0    |
| 2.9162        | 1.64  | 300  | 2.8922          | 1.0    |
| 2.8398        | 2.19  | 400  | 2.3875          | 0.9756 |
| 1.2121        | 2.74  | 500  | 0.8873          | 0.7050 |
| 0.5833        | 3.29  | 600  | 0.6149          | 0.5276 |
| 0.4103        | 3.84  | 700  | 0.5275          | 0.4575 |
| 0.2901        | 4.38  | 800  | 0.5056          | 0.4510 |
| 0.2584        | 4.93  | 900  | 0.5021          | 0.4363 |
| 0.1929        | 5.48  | 1000 | 0.4881          | 0.4087 |
| 0.1747        | 6.03  | 1100 | 0.4961          | 0.4209 |
| 0.1469        | 6.58  | 1200 | 0.4893          | 0.3868 |


### Framework versions

- Transformers 4.28.0
- Pytorch 2.0.1+cu117
- Datasets 2.14.4
- Tokenizers 0.13.3