|
--- |
|
license: apache-2.0 |
|
base_model: openai/whisper-large-v2 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: Whisper-large-Jibbali_lang |
|
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-large-Jibbali_lang |
|
|
|
This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0131 |
|
|
|
## 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: 0.001 |
|
- train_batch_size: 6 |
|
- eval_batch_size: 6 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 50 |
|
- num_epochs: 10 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 0.0224 | 1.0 | 300 | 0.0322 | |
|
| 0.0207 | 2.0 | 600 | 0.0389 | |
|
| 0.0243 | 3.0 | 900 | 0.0349 | |
|
| 0.0032 | 4.0 | 1200 | 0.0174 | |
|
| 0.0044 | 5.0 | 1500 | 0.0146 | |
|
| 0.0066 | 6.0 | 1800 | 0.0132 | |
|
| 0.0033 | 7.0 | 2100 | 0.0141 | |
|
| 0.0017 | 8.0 | 2400 | 0.0118 | |
|
| 0.0008 | 9.0 | 2700 | 0.0130 | |
|
| 0.0015 | 10.0 | 3000 | 0.0131 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.17.0 |
|
- Tokenizers 0.15.2 |
|
|
