File size: 5,405 Bytes
0edebf9 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: wav2vec2LugandaASR
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. -->
# wav2vec2LugandaASR
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6798
- Wer: 0.5291
## 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: 0.0003
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 130
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:------:|:----:|:---------------:|:------:|
| 2.8406 | 1.94 | 100 | 2.8577 | 0.9993 |
| 2.7812 | 3.88 | 200 | 2.8315 | 0.9993 |
| 1.1352 | 5.83 | 300 | 1.0099 | 1.1813 |
| 0.5333 | 7.77 | 400 | 0.5782 | 0.7937 |
| 0.3341 | 9.71 | 500 | 0.5899 | 0.7265 |
| 0.2432 | 11.65 | 600 | 0.5352 | 0.7162 |
| 0.2146 | 13.59 | 700 | 0.5439 | 0.6466 |
| 0.1998 | 15.53 | 800 | 0.5865 | 0.6618 |
| 0.1576 | 17.48 | 900 | 0.5598 | 0.6309 |
| 0.1665 | 19.42 | 1000 | 0.5400 | 0.6135 |
| 0.1191 | 21.36 | 1100 | 0.5496 | 0.6004 |
| 0.1038 | 23.3 | 1200 | 0.6248 | 0.6084 |
| 0.104 | 25.24 | 1300 | 0.5517 | 0.5934 |
| 0.1025 | 27.18 | 1400 | 0.5933 | 0.6008 |
| 0.1024 | 29.13 | 1500 | 0.5693 | 0.5901 |
| 0.0935 | 31.07 | 1600 | 0.5842 | 0.5899 |
| 0.0851 | 33.01 | 1700 | 0.6291 | 0.6086 |
| 0.0773 | 34.95 | 1800 | 0.6138 | 0.5812 |
| 0.0873 | 36.89 | 1900 | 0.5944 | 0.5729 |
| 0.0634 | 38.83 | 2000 | 0.6180 | 0.5807 |
| 0.0631 | 40.78 | 2100 | 0.5904 | 0.5704 |
| 0.0709 | 42.72 | 2200 | 0.5855 | 0.5791 |
| 0.0576 | 44.66 | 2300 | 0.6096 | 0.5789 |
| 0.0605 | 46.6 | 2400 | 0.5749 | 0.5617 |
| 0.0795 | 48.54 | 2500 | 0.5974 | 0.5749 |
| 0.0543 | 50.49 | 2600 | 0.6386 | 0.5754 |
| 0.0531 | 52.43 | 2700 | 0.6469 | 0.5794 |
| 0.0554 | 54.37 | 2800 | 0.6340 | 0.5555 |
| 0.0515 | 56.31 | 2900 | 0.6500 | 0.5762 |
| 0.0439 | 58.25 | 3000 | 0.6376 | 0.5758 |
| 0.0461 | 60.19 | 3100 | 0.6265 | 0.5711 |
| 0.0479 | 62.14 | 3200 | 0.6230 | 0.5707 |
| 0.039 | 64.08 | 3300 | 0.6337 | 0.5584 |
| 0.0397 | 66.02 | 3400 | 0.6347 | 0.5736 |
| 0.0509 | 67.96 | 3500 | 0.5946 | 0.5483 |
| 0.0471 | 69.9 | 3600 | 0.6355 | 0.5584 |
| 0.0481 | 71.84 | 3700 | 0.6514 | 0.5559 |
| 0.0484 | 73.79 | 3800 | 0.6373 | 0.5566 |
| 0.041 | 75.73 | 3900 | 0.6736 | 0.5646 |
| 0.0349 | 77.67 | 4000 | 0.6375 | 0.5622 |
| 0.0349 | 79.61 | 4100 | 0.6158 | 0.5506 |
| 0.0273 | 81.55 | 4200 | 0.6914 | 0.5666 |
| 0.029 | 83.5 | 4300 | 0.6361 | 0.5399 |
| 0.0353 | 85.44 | 4400 | 0.6397 | 0.5584 |
| 0.0289 | 87.38 | 4500 | 0.6554 | 0.5499 |
| 0.0257 | 89.32 | 4600 | 0.6676 | 0.5557 |
| 0.0403 | 91.26 | 4700 | 0.6440 | 0.5584 |
| 0.0361 | 93.2 | 4800 | 0.6587 | 0.5521 |
| 0.0304 | 95.15 | 4900 | 0.6837 | 0.5454 |
| 0.0289 | 97.09 | 5000 | 0.6684 | 0.5370 |
| 0.0282 | 99.03 | 5100 | 0.6556 | 0.5296 |
| 0.0302 | 100.97 | 5200 | 0.6833 | 0.5394 |
| 0.0196 | 102.91 | 5300 | 0.6837 | 0.5291 |
| 0.0255 | 104.85 | 5400 | 0.6644 | 0.5374 |
| 0.0209 | 106.8 | 5500 | 0.6700 | 0.5289 |
| 0.0243 | 108.74 | 5600 | 0.6835 | 0.5338 |
| 0.0203 | 110.68 | 5700 | 0.6850 | 0.5410 |
| 0.0237 | 112.62 | 5800 | 0.6561 | 0.5349 |
| 0.0251 | 114.56 | 5900 | 0.6776 | 0.5298 |
| 0.0177 | 116.5 | 6000 | 0.6748 | 0.5282 |
| 0.0232 | 118.45 | 6100 | 0.6767 | 0.5296 |
| 0.0257 | 120.39 | 6200 | 0.6793 | 0.5320 |
| 0.0194 | 122.33 | 6300 | 0.6804 | 0.5303 |
| 0.0304 | 124.27 | 6400 | 0.6798 | 0.5287 |
| 0.0251 | 126.21 | 6500 | 0.6798 | 0.5291 |
| 0.0201 | 128.16 | 6600 | 0.6798 | 0.5291 |
### Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.0
- Tokenizers 0.13.3
|