Configuration Parsing
Warning:
In adapter_config.json: "peft.task_type" must be a string
Demo Model Whisper Large
This model is a fine-tuned version of EdoAbati/whisper-medium-it on the b-brave/speech_disorders_voice dataset. It achieves the following results on the evaluation set:
- Loss: 0.2483
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: 8
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 64
- num_epochs: 8
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
2.7042 | 0.75 | 48 | 0.4060 |
0.4224 | 1.5 | 96 | 0.3377 |
0.3628 | 2.25 | 144 | 0.3201 |
0.3135 | 3.0 | 192 | 0.2964 |
0.1907 | 3.75 | 240 | 0.2879 |
0.1446 | 4.5 | 288 | 0.3303 |
0.1011 | 5.25 | 336 | 0.2793 |
0.0678 | 6.0 | 384 | 0.2597 |
0.0421 | 6.75 | 432 | 0.2541 |
0.024 | 7.5 | 480 | 0.2483 |
Framework versions
- PEFT 0.11.1
- Transformers 4.42.4
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 4
Model tree for miosipof/asr_michael_test_2.2
Base model
EdoAbati/whisper-medium-it
Adapter
this model