---
library_name: transformers
license: apache-2.0
base_model: openai/whisper-medium
tags:
- generated_from_trainer
datasets:
- bemgen
metrics:
- wer
model-index:
- name: whisper-medium-bemgen-balanced-model
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: bemgen
      type: bemgen
    metrics:
    - name: Wer
      type: wer
      value: 0.4413347685683531
---

<!-- 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-medium-bemgen-balanced-model

This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co./openai/whisper-medium) on the bemgen dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5412
- Wer: 0.4413

## 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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 5000
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Wer    |
|:-------------:|:------:|:----:|:---------------:|:------:|
| 3.7045        | 0.3960 | 200  | 0.9162          | 0.6827 |
| 2.687         | 0.7921 | 400  | 0.6818          | 0.5352 |
| 1.7185        | 1.1881 | 600  | 0.6266          | 0.4988 |
| 1.7232        | 1.5842 | 800  | 0.5674          | 0.4592 |
| 1.6083        | 1.9802 | 1000 | 0.5412          | 0.4413 |
| 0.7643        | 2.3762 | 1200 | 0.5652          | 0.4280 |
| 0.8362        | 2.7723 | 1400 | 0.5455          | 0.4052 |
| 0.422         | 3.1683 | 1600 | 0.5771          | 0.3991 |


### Framework versions

- Transformers 4.47.1
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0