metadata
base_model: openai/whisper-tiny
datasets:
- fleurs
language:
- de
library_name: transformers
license: apache-2.0
metrics:
- wer
tags:
- hf-asr-leaderboard
- generated_from_trainer
model-index:
- name: Whisper Tiny German - Chee Li
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: Google Fleurs
type: fleurs
config: de_de
split: None
args: 'config: de split: test'
metrics:
- type: wer
value: 164.4974887940811
name: Wer
Whisper Tiny German - Chee Li
This model is a fine-tuned version of openai/whisper-tiny on the Google Fleurs dataset. It achieves the following results on the evaluation set:
- Loss: 0.8331
- Wer: 164.4975
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: 8
- seed: 42
- 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: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.1672 | 4.7619 | 1000 | 0.6545 | 118.3939 |
| 0.0219 | 9.5238 | 2000 | 0.7572 | 140.6221 |
| 0.0081 | 14.2857 | 3000 | 0.8154 | 144.2296 |
| 0.0065 | 19.0476 | 4000 | 0.8331 | 164.4975 |
Framework versions
- Transformers 4.46.2
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.20.1