Hugging Face's logo

Neprox
/
STT-swedish-base-model

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Swedish
whisper
hf-asr-leaderboard
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
STT-swedish-base-model / README.md
Neprox's picture
Neprox
update model card README.md
2cbc8b5
preview code
|
raw
history blame contribute delete
1.7 kB
metadata
language:
  - sv
license: apache-2.0
tags:
  - hf-asr-leaderboard
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: Whisper Small - Swedish
    results: []

Whisper Small - Swedish

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3700
  • Wer: 20.2173

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: 4
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 6000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.048 2.59 2000 0.2925 20.3893
0.0045 5.18 4000 0.3258 19.6553
0.0054 6.48 5000 0.3613 20.3087
0.0028 7.77 6000 0.3700 20.2173

Framework versions

  • Transformers 4.25.0.dev0
  • Pytorch 1.12.1
  • Datasets 2.7.1
  • Tokenizers 0.13.2