|
--- |
|
language: |
|
- en |
|
license: apache-2.0 |
|
base_model: openai/whisper-base |
|
tags: |
|
- hf-asr-leaderboard |
|
- generated_from_trainer |
|
metrics: |
|
- wer |
|
model-index: |
|
- name: Whisper Base Medical |
|
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 Base Medical |
|
|
|
This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co./openai/whisper-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2595 |
|
- Wer: 24.0503 |
|
|
|
## 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: 8 |
|
- total_train_batch_size: 32 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 500 |
|
- num_epochs: 5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|:-------------:|:-----:|:----:|:---------------:|:-------:| |
|
| 0.3836 | 1.0 | 184 | 0.5763 | 29.2094 | |
|
| 0.2101 | 2.0 | 368 | 0.3948 | 30.2361 | |
|
| 0.1197 | 3.0 | 552 | 0.3029 | 27.1047 | |
|
| 0.0528 | 4.0 | 737 | 0.2583 | 24.1273 | |
|
| 0.0261 | 4.99 | 920 | 0.2595 | 24.0503 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.33.0 |
|
- Pytorch 2.0.0 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.13.3 |
|
|
