Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Safetensors
whisper
Generated from Trainer
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM
processor = AutoProcessor.from_pretrained("jlvdoorn/whisper-medium.en-atcosim")
model = AutoModelForMultimodalLM.from_pretrained("jlvdoorn/whisper-medium.en-atcosim")Quick Links
whisper-medium.en-atcosim
This model is a fine-tuned version of openai/whisper-medium.en on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0542
- Wer: 1.4169
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: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10
- num_epochs: 100
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2006 | 8.33 | 500 | 0.0410 | 1.2456 |
| 0.0003 | 16.67 | 1000 | 0.0443 | 1.2224 |
| 0.0001 | 25.0 | 1500 | 0.0473 | 1.1854 |
| 0.0 | 33.33 | 2000 | 0.0489 | 1.2039 |
| 0.0 | 41.67 | 2500 | 0.0501 | 1.2224 |
| 0.0 | 50.0 | 3000 | 0.0511 | 1.2873 |
| 0.0 | 58.33 | 3500 | 0.0520 | 1.3012 |
| 0.0 | 66.67 | 4000 | 0.0527 | 1.3197 |
| 0.0 | 75.0 | 4500 | 0.0533 | 1.3891 |
| 0.0 | 83.33 | 5000 | 0.0537 | 1.4077 |
| 0.0 | 91.67 | 5500 | 0.0541 | 1.4169 |
| 0.0 | 100.0 | 6000 | 0.0542 | 1.4169 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2
- Datasets 2.15.0
- Tokenizers 0.15.0
- Downloads last month
- 2
Model tree for jlvdoorn/whisper-medium.en-atcosim
Base model
openai/whisper-medium.en
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="jlvdoorn/whisper-medium.en-atcosim")