File size: 4,563 Bytes
9f27d33 64ed143 9f27d33 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-large-v2-atcosim
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-atcosim
This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the [ATCOSIM](https://huggingface.co/datasets/jlvdoorn/ATCOSIM) dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0679
- Wer: 6.2234
## 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
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- training_steps: 50000
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:------:|:-----:|:---------------:|:-------:|
| 0.0147 | 2.09 | 1000 | 0.0373 | 4.7972 |
| 0.0058 | 4.18 | 2000 | 0.0379 | 4.5934 |
| 0.0012 | 6.28 | 3000 | 0.0388 | 4.5425 |
| 0.0067 | 8.37 | 4000 | 0.0382 | 4.0470 |
| 0.0026 | 10.46 | 5000 | 0.0382 | 2.9959 |
| 0.0016 | 12.55 | 6000 | 0.0457 | 2.9496 |
| 0.0027 | 14.64 | 7000 | 0.0473 | 4.6305 |
| 0.001 | 16.74 | 8000 | 0.0419 | 3.2969 |
| 0.0011 | 18.83 | 9000 | 0.0424 | 4.4592 |
| 0.0013 | 20.92 | 10000 | 0.0432 | 5.8807 |
| 0.0019 | 23.01 | 11000 | 0.0454 | 3.7646 |
| 0.0004 | 25.1 | 12000 | 0.0443 | 9.5110 |
| 0.004 | 27.2 | 13000 | 0.0442 | 2.8385 |
| 0.0018 | 29.29 | 14000 | 0.0444 | 2.5282 |
| 0.0011 | 31.38 | 15000 | 0.0467 | 4.0980 |
| 0.0002 | 33.47 | 16000 | 0.0469 | 3.9128 |
| 0.003 | 35.56 | 17000 | 0.0454 | 4.7462 |
| 0.0001 | 37.66 | 18000 | 0.0459 | 3.1950 |
| 0.0006 | 39.75 | 19000 | 0.0451 | 2.6579 |
| 0.0014 | 41.84 | 20000 | 0.0464 | 1.6855 |
| 0.0 | 43.93 | 21000 | 0.0487 | 2.3106 |
| 0.0005 | 46.03 | 22000 | 0.0535 | 7.3717 |
| 0.0001 | 48.12 | 23000 | 0.0482 | 6.9411 |
| 0.0002 | 50.21 | 24000 | 0.0484 | 13.0580 |
| 0.0001 | 52.3 | 25000 | 0.0481 | 18.0219 |
| 0.0 | 54.39 | 26000 | 0.0523 | 14.7342 |
| 0.0 | 56.49 | 27000 | 0.0552 | 11.1132 |
| 0.0004 | 58.58 | 28000 | 0.0521 | 2.5190 |
| 0.0001 | 60.67 | 29000 | 0.0490 | 4.4036 |
| 0.0 | 62.76 | 30000 | 0.0497 | 2.8246 |
| 0.0 | 64.85 | 31000 | 0.0513 | 2.8755 |
| 0.0 | 66.95 | 32000 | 0.0526 | 2.9172 |
| 0.0 | 69.04 | 33000 | 0.0539 | 3.0098 |
| 0.0 | 71.13 | 34000 | 0.0552 | 3.0144 |
| 0.0 | 73.22 | 35000 | 0.0566 | 3.1209 |
| 0.0 | 75.31 | 36000 | 0.0580 | 3.2321 |
| 0.0 | 77.41 | 37000 | 0.0594 | 3.4729 |
| 0.0 | 79.5 | 38000 | 0.0607 | 3.6164 |
| 0.0 | 81.59 | 39000 | 0.0620 | 3.9035 |
| 0.0 | 83.68 | 40000 | 0.0632 | 4.0656 |
| 0.0 | 85.77 | 41000 | 0.0642 | 4.3202 |
| 0.0 | 87.87 | 42000 | 0.0651 | 4.4453 |
| 0.0 | 89.96 | 43000 | 0.0659 | 4.9361 |
| 0.0 | 92.05 | 44000 | 0.0664 | 5.2186 |
| 0.0 | 94.14 | 45000 | 0.0670 | 5.6029 |
| 0.0 | 96.23 | 46000 | 0.0673 | 5.7835 |
| 0.0 | 98.33 | 47000 | 0.0676 | 6.0520 |
| 0.0 | 100.42 | 48000 | 0.0678 | 6.1122 |
| 0.0 | 102.51 | 49000 | 0.0679 | 6.2141 |
| 0.0 | 104.6 | 50000 | 0.0679 | 6.2234 |
### Framework versions
- Transformers 4.30.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3
|