# Load model directly
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
processor = AutoProcessor.from_pretrained("lalok/nectar_aihub_model_10000steps")
model = AutoModelForSpeechSeq2Seq.from_pretrained("lalok/nectar_aihub_model_10000steps", device_map="auto")Quick Links
nectar_aihub_model_10000steps
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: 0.1407
- Cer: 10.7869
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: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 10000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 0.2001 | 0.1743 | 2000 | 0.1952 | 13.8677 |
| 0.1786 | 0.3486 | 4000 | 0.1752 | 11.6286 |
| 0.1438 | 0.5229 | 6000 | 0.1595 | 11.4573 |
| 0.1521 | 0.6972 | 8000 | 0.1470 | 10.8939 |
| 0.1396 | 0.8715 | 10000 | 0.1407 | 10.7869 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.2.2+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1
- Downloads last month
- 2
Model tree for lalok/nectar_aihub_model_10000steps
Base model
openai/whisper-medium
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="lalok/nectar_aihub_model_10000steps")