metadata
base_model: openai/whisper-medium
datasets:
- b-brave/speech_disorders_voice
language:
- it
library_name: peft
license: apache-2.0
metrics:
- wer
tags:
- generated_from_trainer
model-index:
- name: Whisper Large v3
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: b-brave/speech_disorders_voice
type: b-brave/speech_disorders_voice
config: default
split: train
args: default
metrics:
- type: wer
value: 13.701431492842536
name: Wer
Whisper Large v3
This model is a fine-tuned version of openai/whisper-medium on the b-brave/speech_disorders_voice dataset. It achieves the following results on the evaluation set:
- Loss: 0.3056
- Wer: 13.7014
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: 32
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 128
- training_steps: 384
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.6048 | 0.9481 | 64 | 0.5505 | 22.0859 |
| 0.2149 | 1.8963 | 128 | 0.3104 | 67.8937 |
| 0.0653 | 2.8444 | 192 | 0.3124 | 15.5419 |
| 0.0255 | 3.7926 | 256 | 0.3050 | 15.1329 |
| 0.013 | 4.7407 | 320 | 0.3051 | 14.1104 |
| 0.009 | 5.6889 | 384 | 0.3056 | 13.7014 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.2.0
- Datasets 2.21.0
- Tokenizers 0.19.1