metadata
license: apache-2.0
base_model: openai/whisper-large-v3
tags:
- generated_from_trainer
datasets:
- honzapucalek/hc_impaired_all_v3
metrics:
- wer
model-index:
- name: hc-impaired-all-v3
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: honzapucalek/hc_impaired_all_v3 cs
type: honzapucalek/hc_impaired_all_v3
config: cs
split: test
args: cs
metrics:
- name: Wer
type: wer
value: 0.11072981011210249
hc-impaired-all-v3
This model is a fine-tuned version of openai/whisper-large-v3 on the honzapucalek/hc_impaired_all_v3 cs dataset. It achieves the following results on the evaluation set:
- Loss: 0.3837
- Wer: 0.1107
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 |
|---|---|---|---|---|
| 0.0275 | 6.87 | 1000 | 0.2212 | 0.1163 |
| 0.0021 | 13.75 | 2000 | 0.3051 | 0.1123 |
| 0.0004 | 20.62 | 3000 | 0.3517 | 0.1113 |
| 0.0001 | 27.49 | 4000 | 0.3760 | 0.1104 |
| 0.0001 | 34.36 | 5000 | 0.3837 | 0.1107 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1