metadata
library_name: transformers
license: apache-2.0
base_model: openai/whisper-base
tags:
- generated_from_trainer
datasets:
- generator
metrics:
- wer
model-index:
- name: whisper-base_ro-80mel
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: generator
type: generator
config: default
split: None
args: default
metrics:
- name: Wer
type: wer
value: 2.4081
whisper-base_ro-80mel
This model is a fine-tuned version of openai/whisper-base on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 0.9904
- Wer: 2.4081
- Cer: 2.9115
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 9
- total_train_batch_size: 144
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1400
- num_epochs: 6
- label_smoothing_factor: 0.05
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 1.114 | 1.0424 | 1500 | 0.9904 | 2.9511 | 3.1317 |
| 0.9647 | 2.0848 | 3000 | 0.8996 | 4.6434 | 4.1911 |
| 0.916 | 3.1272 | 4500 | 0.8811 | 5.4089 | 4.8542 |
| 0.8868 | 4.1696 | 6000 | 0.8595 | 4.9368 | 4.6589 |
Framework versions
- Transformers 4.57.0
- Pytorch 2.9.1+cu128
- Datasets 4.4.1
- Tokenizers 0.22.1