metadata
license: apache-2.0
base_model: openai/whisper-large-v3
tags:
- generated_from_trainer
datasets:
- honzapucalek/hc_train_v3_independent_v2
metrics:
- wer
model-index:
- name: hc-train-v3-independent-v2
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: honzapucalek/hc_train_v3_independent_v2 cs
type: honzapucalek/hc_train_v3_independent_v2
config: cs
split: test
args: cs
metrics:
- name: Wer
type: wer
value: 0.1169068862960421
hc-train-v3-independent-v2
This model is a fine-tuned version of openai/whisper-large-v3 on the honzapucalek/hc_train_v3_independent_v2 cs dataset. It achieves the following results on the evaluation set:
- Loss: 0.3728
- Wer: 0.1169
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- 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: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0079 | 13.51 | 1000 | 0.2854 | 0.1256 |
| 0.0037 | 27.03 | 2000 | 0.3198 | 0.1373 |
| 0.0002 | 40.54 | 3000 | 0.3459 | 0.1177 |
| 0.0001 | 54.05 | 4000 | 0.3650 | 0.1168 |
| 0.0001 | 67.57 | 5000 | 0.3728 | 0.1169 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1