metadata
library_name: transformers
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: Whisper Small superU
results: []
Whisper Small superU
This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 6.3374
- Wer: 57.2770
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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 128
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0 | 100.0 | 100 | 2.1912 | 59.6244 |
| 0.0 | 200.0 | 200 | 2.3111 | 58.2160 |
| 0.0 | 300.0 | 300 | 2.3886 | 58.2160 |
| 0.0 | 400.0 | 400 | 2.5319 | 60.5634 |
| 0.0 | 500.0 | 500 | 2.7160 | 60.0939 |
| 0.0 | 600.0 | 600 | 2.9609 | 61.0329 |
| 0.0 | 700.0 | 700 | 3.2141 | 61.0329 |
| 0.0 | 800.0 | 800 | 3.4591 | 61.5023 |
| 0.0 | 900.0 | 900 | 3.7213 | 62.4413 |
| 0.0 | 1000.0 | 1000 | 3.9804 | 61.0329 |
| 0.0 | 1100.0 | 1100 | 4.2234 | 78.4038 |
| 0.0 | 1200.0 | 1200 | 4.4138 | 63.3803 |
| 0.0 | 1300.0 | 1300 | 4.5889 | 77.9343 |
| 0.0 | 1400.0 | 1400 | 4.7946 | 70.8920 |
| 0.0 | 1500.0 | 1500 | 4.9337 | 65.7277 |
| 0.0 | 1600.0 | 1600 | 5.0758 | 56.8075 |
| 0.0 | 1700.0 | 1700 | 5.2692 | 56.8075 |
| 0.0 | 1800.0 | 1800 | 5.4087 | 56.8075 |
| 0.0 | 1900.0 | 1900 | 5.5500 | 56.8075 |
| 0.0 | 2000.0 | 2000 | 5.6783 | 56.8075 |
| 0.0 | 2100.0 | 2100 | 5.6287 | 56.8075 |
| 0.0 | 2200.0 | 2200 | 5.6852 | 56.3380 |
| 0.0 | 2300.0 | 2300 | 5.7374 | 56.3380 |
| 0.0 | 2400.0 | 2400 | 5.8023 | 56.3380 |
| 0.0 | 2500.0 | 2500 | 5.8672 | 57.2770 |
| 0.0 | 2600.0 | 2600 | 5.9427 | 57.2770 |
| 0.0 | 2700.0 | 2700 | 5.9891 | 57.2770 |
| 0.0 | 2800.0 | 2800 | 6.0490 | 57.2770 |
| 0.0 | 2900.0 | 2900 | 6.0639 | 57.2770 |
| 0.0 | 3000.0 | 3000 | 6.1095 | 57.2770 |
| 0.0 | 3100.0 | 3100 | 6.1477 | 57.2770 |
| 0.0 | 3200.0 | 3200 | 6.2039 | 57.2770 |
| 0.0 | 3300.0 | 3300 | 6.2346 | 57.2770 |
| 0.0 | 3400.0 | 3400 | 6.2567 | 57.2770 |
| 0.0 | 3500.0 | 3500 | 6.2841 | 57.2770 |
| 0.0 | 3600.0 | 3600 | 6.3028 | 57.2770 |
| 0.0 | 3700.0 | 3700 | 6.3029 | 57.2770 |
| 0.0 | 3800.0 | 3800 | 6.3294 | 57.2770 |
| 0.0 | 3900.0 | 3900 | 6.3346 | 57.2770 |
| 0.0 | 4000.0 | 4000 | 6.3374 | 57.2770 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1