README.md

---
library_name: transformers
language:
- nl
license: apache-2.0
base_model: openai/whisper-large-v2
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: Whisper Large V2
  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 Large V2

This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3623
- Wer: 21.2197

## 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: 3e-05
- train_batch_size: 12
- 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: 20
- num_epochs: 5

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer     |
|:-------------:|:-----:|:----:|:---------------:|:-------:|
| 0.5651        | 0.75  | 15   | 0.4091          | 25.8916 |
| 0.2755        | 1.5   | 30   | 0.3437          | 23.1455 |
| 0.1284        | 2.25  | 45   | 0.3333          | 19.0086 |
| 0.0768        | 3.0   | 60   | 0.3305          | 27.7461 |
| 0.0365        | 3.75  | 75   | 0.3449          | 24.8752 |
| 0.0224        | 4.5   | 90   | 0.3623          | 21.2197 |


### Framework versions

- Transformers 4.45.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1