---
library_name: transformers
license: apache-2.0
base_model: openai/whisper-large-v2
tags:
- generated_from_trainer
datasets:
- common_voice_17_0
metrics:
- wer
model-index:
- name: asr-whisper-helpline-sw-v1
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: common_voice_17_0
      type: common_voice_17_0
      config: sw
      split: validation
      args: sw
    metrics:
    - name: Wer
      type: wer
      value: 23.61977361977362
---

<!-- 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. -->

# asr-whisper-helpline-sw-v1

This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the common_voice_17_0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4024
- Wer: 23.6198

## 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
- optimizer: Use OptimizerNames.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: 10000
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step  | Validation Loss | Wer     |
|:-------------:|:------:|:-----:|:---------------:|:-------:|
| 0.5275        | 0.05   | 500   | 0.7083          | 43.6821 |
| 0.4754        | 0.1    | 1000  | 0.5733          | 42.0651 |
| 0.2844        | 0.15   | 1500  | 0.5390          | 36.2093 |
| 0.2647        | 0.2    | 2000  | 0.5060          | 34.3151 |
| 0.2939        | 0.25   | 2500  | 0.4761          | 34.8926 |
| 0.285         | 1.0094 | 3000  | 0.4610          | 35.0427 |
| 0.1837        | 1.0594 | 3500  | 0.4516          | 34.0725 |
| 0.2148        | 1.1094 | 4000  | 0.4309          | 33.1023 |
| 0.2611        | 1.1594 | 4500  | 0.4328          | 29.0945 |
| 0.1923        | 1.2094 | 5000  | 0.4180          | 31.2428 |
| 0.0849        | 1.2594 | 5500  | 0.4229          | 25.8027 |
| 0.1913        | 2.0188 | 6000  | 0.4051          | 28.4592 |
| 0.0693        | 2.0688 | 6500  | 0.4256          | 29.1753 |
| 0.1261        | 2.1188 | 7000  | 0.4057          | 28.8750 |
| 0.0808        | 2.1688 | 7500  | 0.4054          | 28.3899 |
| 0.0524        | 2.2188 | 8000  | 0.4248          | 25.6757 |
| 0.0845        | 2.2688 | 8500  | 0.4159          | 26.5073 |
| 0.0611        | 3.0282 | 9000  | 0.4142          | 25.4447 |
| 0.0562        | 3.0782 | 9500  | 0.4104          | 25.6872 |
| 0.0867        | 3.1282 | 10000 | 0.4024          | 23.6198 |


### Framework versions

- Transformers 4.56.2
- Pytorch 2.8.0+cu128
- Datasets 2.21.0
- Tokenizers 0.22.1