|
--- |
|
library_name: transformers |
|
license: apache-2.0 |
|
base_model: shreyasdesaisuperU/whisper-medium-attempt2-1000-orders |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- wer |
|
model-index: |
|
- name: Whisper Medium 1000 orders Eleven Labs SSD superU |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# Whisper Medium 1000 orders Eleven Labs SSD superU |
|
|
|
This model is a fine-tuned version of [shreyasdesaisuperU/whisper-medium-attempt2-1000-orders](https://huggingface.co/shreyasdesaisuperU/whisper-medium-attempt2-1000-orders) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0128 |
|
- Wer: 0.8606 |
|
|
|
## 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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 500 |
|
- training_steps: 2000 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|:-------------:|:------:|:----:|:---------------:|:-------:| |
|
| 0.0668 | 0.4032 | 100 | 0.0388 | 17.3838 | |
|
| 0.0142 | 0.8065 | 200 | 0.0061 | 11.3597 | |
|
| 0.0075 | 1.2097 | 300 | 0.0075 | 9.6386 | |
|
| 0.0073 | 1.6129 | 400 | 0.0104 | 7.7453 | |
|
| 0.0087 | 2.0161 | 500 | 0.0125 | 2.9260 | |
|
| 0.0046 | 2.4194 | 600 | 0.0080 | 1.5491 | |
|
| 0.0087 | 2.8226 | 700 | 0.0039 | 1.7212 | |
|
| 0.0066 | 3.2258 | 800 | 0.0042 | 1.3769 | |
|
| 0.0032 | 3.6290 | 900 | 0.0095 | 1.0327 | |
|
| 0.0027 | 4.0323 | 1000 | 0.0114 | 1.5491 | |
|
| 0.0021 | 4.4355 | 1100 | 0.0099 | 1.7212 | |
|
| 0.0039 | 4.8387 | 1200 | 0.0121 | 1.8933 | |
|
| 0.0017 | 5.2419 | 1300 | 0.0126 | 1.3769 | |
|
| 0.0033 | 5.6452 | 1400 | 0.0093 | 1.8933 | |
|
| 0.0037 | 6.0484 | 1500 | 0.0126 | 1.2048 | |
|
| 0.0013 | 6.4516 | 1600 | 0.0090 | 1.2048 | |
|
| 0.0014 | 6.8548 | 1700 | 0.0102 | 1.2048 | |
|
| 0.0002 | 7.2581 | 1800 | 0.0115 | 0.8606 | |
|
| 0.0005 | 7.6613 | 1900 | 0.0142 | 1.0327 | |
|
| 0.0002 | 8.0645 | 2000 | 0.0128 | 0.8606 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.46.2 |
|
- Pytorch 2.2.2+cu121 |
|
- Datasets 3.1.0 |
|
- Tokenizers 0.20.3 |
|
|
