---
language:
- sv
- 'no'
- da
license: apache-2.0
tags:
- whisper-event
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_11_0
- babelbox/babelbox_voice
- NbAiLab/NST
- NbAiLab/NPSC
- google/fleurs
metrics:
- wer
model-index:
- name: Whisper Medium Nordic
  results:
  - task:
      type: automatic-speech-recognition
      name: Automatic Speech Recognition
    dataset:
      name: mozilla-foundation/common_voice_11_0
      type: mozilla-foundation/common_voice_11_0
      config: sv-SE
      split: test
    metrics:
    - name: Wer
      type: wer
      value: 11.31
  - task:
      type: automatic-speech-recognition
      name: Automatic Speech Recognition
    dataset:
      name: mozilla-foundation/common_voice_11_0
      type: mozilla-foundation/common_voice_11_0
      config: da
      split: test
    metrics:
    - name: Wer
      type: wer
      value: 14.86
  - task:
      type: automatic-speech-recognition
      name: Automatic Speech Recognition
    dataset:
      name: mozilla-foundation/common_voice_11_0
      type: mozilla-foundation/common_voice_11_0
      config: nn-NO
      split: test
    metrics:
    - name: Wer
      type: wer
      value: 37.02
  

      
---

<!-- 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 Nordic

This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the [mozilla-foundation/common_voice_11_0](https://huggingface.co/datasets/mozilla-foundation/common_voice_11_0) (sv-SE, da, nn-NO), the [babelbox/babelbox_voice](https://huggingface.co/datasets/babelbox/babelbox_voice) (Swedish radio), the [NbAiLab/NST](https://huggingface.co/datasets/NbAiLab/NST) (Norwegian radio), the [NbAiLab/NPSC](https://huggingface.co/datasets/NbAiLab/NPSC) (Norwegian parliament) and the [google/fleurs](https://huggingface.co/datasets/google/fleurs) (sv_se, da_dk, nb_no) datasets. The goal is to leverage transfer learning across Nordic languages, which have strong similarities.

It achieves the following results on the common voice Swedish test set:
- Loss: 0.2129
- Wer: 11.3079

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

Please note that a bug during training prevented us from evaluating WER correctly.
Validation loss suggests we started overfitting after 5000/6000 steps.

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 3e-06
- train_batch_size: 32
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 10000
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Wer     |
|:-------------:|:-----:|:-----:|:---------------:|:-------:|
| 0.3056        | 0.1   | 1000  | 0.2670          | 99.9221 |
| 0.16          | 0.2   | 2000  | 0.2322          | 99.6640 |
| 0.1309        | 0.3   | 3000  | 0.2152          | 98.9759 |
| 0.097         | 0.4   | 4000  | 0.2112          | 100.0   |
| 0.091         | 0.5   | 5000  | 0.2094          | 99.7312 |
| 0.1098        | 0.6   | 6000  | 0.2098          | 98.6077 |
| 0.0637        | 0.7   | 7000  | 0.2148          | 98.4625 |
| 0.0718        | 0.8   | 8000  | 0.2151          | 99.8710 |
| 0.0517        | 0.9   | 9000  | 0.2175          | 97.2342 |
| 0.0465        | 1.0   | 10000 | 0.2129          | 96.3552 |


### Framework versions

- Transformers 4.26.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2