---
language:
- de
license: apache-2.0
tags:
- sbb-asr
- generated_from_trainer
datasets:
- marccgrau/sbbdata_allSNR
metrics:
- wer
model-index:
- name: Whisper Small German SBB all SNR - v4
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: SBB Dataset 05.01.2023
      type: marccgrau/sbbdata_allSNR
      args: 'config: German, split: train, test, val'
    metrics:
    - name: Wer
      type: wer
      value: 0.02219403931515536
---

<!-- 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 Small German SBB all SNR - v4

This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the SBB Dataset 05.01.2023 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0287
- Wer: 0.0222

## 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: 5e-06
- train_batch_size: 64
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- training_steps: 700
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 1.6894        | 0.71  | 100  | 0.4702          | 0.4661 |
| 0.1896        | 1.42  | 200  | 0.0322          | 0.0241 |
| 0.0297        | 2.13  | 300  | 0.0349          | 0.0228 |
| 0.0181        | 2.84  | 400  | 0.0250          | 0.0209 |
| 0.0154        | 3.55  | 500  | 0.0298          | 0.0209 |
| 0.0112        | 4.26  | 600  | 0.0327          | 0.0222 |
| 0.009         | 4.96  | 700  | 0.0287          | 0.0222 |


### Framework versions

- Transformers 4.25.1
- Pytorch 1.13.1
- Datasets 2.8.0
- Tokenizers 0.12.1