# Load model directly
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
processor = AutoProcessor.from_pretrained("cladsu/prueba-medium-whisper")
model = AutoModelForSpeechSeq2Seq.from_pretrained("cladsu/prueba-medium-whisper")Quick Links
prueba-medium-whisper
This model is a fine-tuned version of openai/whisper-medium on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.5431
- Wer: 92.5913
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1
- training_steps: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.0572 | 0.0091 | 5 | 1.9552 | 79.3737 |
| 1.7179 | 0.0182 | 10 | 1.6960 | 112.0115 |
| 1.4811 | 0.0273 | 15 | 1.5831 | 91.7683 |
| 1.513 | 0.0364 | 20 | 1.5431 | 92.5913 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.0
- Tokenizers 0.19.1
- Downloads last month
- 3
Model tree for cladsu/prueba-medium-whisper
Base model
openai/whisper-medium
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="cladsu/prueba-medium-whisper")