metadata
library_name: peft
language:
- it
license: apache-2.0
base_model: openai/whisper-medium
tags:
- generated_from_trainer
datasets:
- Dysarthria_Synthetic_Easycall_Common
metrics:
- wer
model-index:
- name: Whisper Medium
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: Dysarthria_Synthetic_Easycall_Common
type: Dysarthria_Synthetic_Easycall_Common
config: default
split: train
args: default
metrics:
- type: wer
value: 81.93548387096774
name: Wer
Whisper Medium
This model is a fine-tuned version of openai/whisper-medium on the Dysarthria_Synthetic_Easycall_Common dataset. It achieves the following results on the evaluation set:
- Loss: 3.7887
- Wer: 81.9355
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
- gradient_accumulation_steps: 2
- 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: 100
- num_epochs: 7
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 6.7033 | 0.6897 | 50 | 6.6396 | 205.1613 |
| 6.091 | 1.3793 | 100 | 5.9043 | 160.0 |
| 5.0041 | 2.0690 | 150 | 5.4640 | 163.2258 |
| 4.4908 | 2.7586 | 200 | 5.0968 | 90.6452 |
| 4.023 | 3.4483 | 250 | 4.7181 | 85.8065 |
| 3.6629 | 4.1379 | 300 | 4.2892 | 83.2258 |
| 3.3164 | 4.8276 | 350 | 4.0228 | 80.9677 |
| 3.2009 | 5.5172 | 400 | 3.8851 | 82.2581 |
| 3.0631 | 6.2069 | 450 | 3.8146 | 82.9032 |
| 3.0098 | 6.8966 | 500 | 3.7887 | 81.9355 |
Framework versions
- PEFT 0.14.0
- Transformers 4.45.2
- Pytorch 2.2.0
- Datasets 3.2.0
- Tokenizers 0.20.3