metadata
language:
- multilingual
license: apache-2.0
base_model: openai/whisper-medium
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_11_0
metrics:
- wer
model-index:
- name: model trenovan na en_de_en simi setu, nastaveni jazyka en overeni3
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: odpovidajici nazvu modelu
type: mozilla-foundation/common_voice_11_0
args: 'config: ende, split: train'
metrics:
- name: Wer
type: wer
value: 31.315296008572197
model trenovan na en_de_en simi setu, nastaveni jazyka en overeni3
This model is a fine-tuned version of openai/whisper-medium on the odpovidajici nazvu modelu dataset. It achieves the following results on the evaluation set:
- Loss: 0.2841
- Wer: 31.3153
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: 6.600000000000001e-06
- 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: 1
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.4886 | 0.1 | 100 | 0.3900 | 12.3225 |
| 0.2534 | 0.2 | 200 | 0.2826 | 15.7514 |
| 0.1143 | 1.05 | 300 | 0.2457 | 16.4211 |
| 0.1134 | 1.15 | 400 | 0.2496 | 14.2245 |
| 0.1186 | 1.25 | 500 | 0.2511 | 13.5280 |
| 0.054 | 2.1 | 600 | 0.2694 | 15.6175 |
| 0.0449 | 2.2 | 700 | 0.2743 | 21.0019 |
| 0.0224 | 3.05 | 800 | 0.2812 | 24.1897 |
| 0.0241 | 3.15 | 900 | 0.2828 | 31.6368 |
| 0.0324 | 3.25 | 1000 | 0.2841 | 31.3153 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.15.2