---
library_name: transformers
license: apache-2.0
base_model: openai/whisper-large-v3
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: multilingual-whisper-v3-scripted
  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. -->

# multilingual-whisper-v3-scripted

This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4568
- Wer: 0.5008
- Cer: 0.1871

## 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-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- training_steps: 2000

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    | Cer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
| 0.8035        | 0.25  | 500  | 0.6861          | 0.6772 | 0.2742 |
| 0.427         | 0.5   | 1000 | 0.5404          | 0.4711 | 0.1497 |
| 0.6455        | 0.75  | 1500 | 0.4845          | 0.5254 | 0.2012 |
| 0.2086        | 1.0   | 2000 | 0.4568          | 0.5008 | 0.1871 |


### Framework versions

- Transformers 4.52.0
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.4