metadata
language:
- de
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_16_0
metrics:
- wer
model-index:
- name: Whisper Tiny CV de
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Common Voice 11.0 de 5%
type: mozilla-foundation/common_voice_16_0
config: de
split: None
args: 'config: de, split: test'
metrics:
- name: Wer
type: wer
value: 72.91819291819291
Whisper Tiny CV de
This model is a fine-tuned version of openai/whisper-tiny on the Common Voice 11.0 de 5% dataset. It achieves the following results on the evaluation set:
- Loss: 0.7117
- Wer: 72.9182
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: 1.35e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 250
- training_steps: 2000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.6076 | 0.2252 | 250 | 0.8347 | 76.3126 |
| 0.5955 | 0.4505 | 500 | 0.7893 | 79.1697 |
| 0.5179 | 0.6757 | 750 | 0.7593 | 82.1978 |
| 0.5189 | 0.9009 | 1000 | 0.7370 | 73.0159 |
| 0.3644 | 1.1261 | 1250 | 0.7254 | 84.1270 |
| 0.394 | 1.3514 | 1500 | 0.7183 | 73.4066 |
| 0.3672 | 1.5766 | 1750 | 0.7152 | 73.1136 |
| 0.3751 | 1.8018 | 2000 | 0.7117 | 72.9182 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1