|
--- |
|
language: |
|
- nl |
|
license: apache-2.0 |
|
base_model: openai/whisper-large-v2 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- wer |
|
model-index: |
|
- name: Whisper Large V2 |
|
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 Large V2 |
|
|
|
This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3047 |
|
- Wer: 10.4756 |
|
|
|
## 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: 3e-05 |
|
- train_batch_size: 12 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 20 |
|
- num_epochs: 5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|:-------------:|:-----:|:----:|:---------------:|:-------:| |
|
| 0.5862 | 0.09 | 30 | 0.3770 | 15.4837 | |
|
| 0.3186 | 0.19 | 60 | 0.3302 | 13.7743 | |
|
| 0.2867 | 0.28 | 90 | 0.3126 | 13.5958 | |
|
| 0.288 | 0.38 | 120 | 0.2984 | 12.1001 | |
|
| 0.2647 | 0.47 | 150 | 0.2963 | 14.9480 | |
|
| 0.2578 | 0.57 | 180 | 0.2984 | 13.6251 | |
|
| 0.2943 | 0.66 | 210 | 0.2910 | 15.0124 | |
|
| 0.2584 | 0.76 | 240 | 0.2758 | 14.6729 | |
|
| 0.2741 | 0.85 | 270 | 0.2724 | 11.9040 | |
|
| 0.2595 | 0.95 | 300 | 0.2743 | 14.1753 | |
|
| 0.2164 | 1.04 | 330 | 0.2688 | 12.1469 | |
|
| 0.1197 | 1.14 | 360 | 0.2665 | 12.0006 | |
|
| 0.1275 | 1.23 | 390 | 0.2690 | 11.4035 | |
|
| 0.1342 | 1.33 | 420 | 0.2742 | 12.2025 | |
|
| 0.1271 | 1.42 | 450 | 0.2695 | 12.0972 | |
|
| 0.1335 | 1.52 | 480 | 0.2728 | 11.3508 | |
|
| 0.1385 | 1.61 | 510 | 0.2669 | 11.5908 | |
|
| 0.1326 | 1.71 | 540 | 0.2631 | 11.8045 | |
|
| 0.1245 | 1.8 | 570 | 0.2621 | 12.0884 | |
|
| 0.1232 | 1.9 | 600 | 0.2597 | 11.6611 | |
|
| 0.1325 | 1.99 | 630 | 0.2576 | 11.6054 | |
|
| 0.0615 | 2.09 | 660 | 0.2724 | 12.8055 | |
|
| 0.0615 | 2.18 | 690 | 0.2703 | 12.1908 | |
|
| 0.0575 | 2.28 | 720 | 0.2699 | 12.0474 | |
|
| 0.0568 | 2.37 | 750 | 0.2722 | 11.8425 | |
|
| 0.0562 | 2.47 | 780 | 0.2734 | 12.9987 | |
|
| 0.0568 | 2.56 | 810 | 0.2696 | 11.2630 | |
|
| 0.0567 | 2.66 | 840 | 0.2749 | 10.9557 | |
|
| 0.058 | 2.75 | 870 | 0.2783 | 11.6025 | |
|
| 0.0608 | 2.85 | 900 | 0.2733 | 11.1605 | |
|
| 0.0586 | 2.94 | 930 | 0.2678 | 11.9830 | |
|
| 0.044 | 3.04 | 960 | 0.2753 | 11.2601 | |
|
| 0.0236 | 3.13 | 990 | 0.2814 | 10.8825 | |
|
| 0.0235 | 3.23 | 1020 | 0.2853 | 11.0376 | |
|
| 0.0229 | 3.32 | 1050 | 0.2865 | 10.7654 | |
|
| 0.0217 | 3.42 | 1080 | 0.2848 | 10.6776 | |
|
| 0.0233 | 3.51 | 1110 | 0.2838 | 10.6600 | |
|
| 0.0223 | 3.61 | 1140 | 0.2867 | 10.6981 | |
|
| 0.0208 | 3.7 | 1170 | 0.2791 | 10.3761 | |
|
| 0.0195 | 3.8 | 1200 | 0.2832 | 10.5020 | |
|
| 0.02 | 3.89 | 1230 | 0.2841 | 10.9176 | |
|
| 0.0204 | 3.99 | 1260 | 0.2817 | 10.4610 | |
|
| 0.0092 | 4.08 | 1290 | 0.2933 | 10.5312 | |
|
| 0.0078 | 4.18 | 1320 | 0.2992 | 10.4727 | |
|
| 0.0068 | 4.27 | 1350 | 0.3026 | 10.3264 | |
|
| 0.0076 | 4.37 | 1380 | 0.3064 | 10.7361 | |
|
| 0.0077 | 4.46 | 1410 | 0.3070 | 10.5752 | |
|
| 0.0073 | 4.56 | 1440 | 0.3070 | 10.5459 | |
|
| 0.0078 | 4.65 | 1470 | 0.3053 | 10.5254 | |
|
| 0.0083 | 4.75 | 1500 | 0.3035 | 10.4317 | |
|
| 0.009 | 4.84 | 1530 | 0.3042 | 10.4669 | |
|
| 0.0074 | 4.94 | 1560 | 0.3047 | 10.4756 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.0.dev0 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.17.0 |
|
- Tokenizers 0.15.0 |
|
|