File size: 4,331 Bytes
c5e4388 481740c c5e4388 481740c |
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 |
---
library_name: transformers
license: apache-2.0
base_model: facebook/wav2vec2-large-xlsr-53
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: xlsr-aiish-no
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. -->
# xlsr-aiish-no
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: 0.0000
- Wer: 0.3093
## 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: 0.0004
- train_batch_size: 8
- eval_batch_size: 8
- 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: 132
- num_epochs: 100
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-------:|:----:|:---------------:|:------:|
| 4.5121 | 2.1505 | 200 | 2.5138 | 1.0 |
| 1.2964 | 4.3011 | 400 | 0.1073 | 0.4756 |
| 0.1795 | 6.4516 | 600 | 0.0381 | 0.3814 |
| 0.096 | 8.6022 | 800 | 0.0067 | 0.3117 |
| 0.0584 | 10.7527 | 1000 | 0.0140 | 0.3227 |
| 0.0457 | 12.9032 | 1200 | 0.0039 | 0.3130 |
| 0.0412 | 15.0538 | 1400 | 0.0024 | 0.3081 |
| 0.0269 | 17.2043 | 1600 | 0.0142 | 0.3093 |
| 0.0276 | 19.3548 | 1800 | 0.0013 | 0.3068 |
| 0.0279 | 21.5054 | 2000 | 0.0044 | 0.3117 |
| 0.0243 | 23.6559 | 2200 | 0.0026 | 0.3105 |
| 0.0178 | 25.8065 | 2400 | 0.0006 | 0.3081 |
| 0.0193 | 27.9570 | 2600 | 0.0115 | 0.3215 |
| 0.0237 | 30.1075 | 2800 | 0.0008 | 0.3068 |
| 0.0146 | 32.2581 | 3000 | 0.0011 | 0.3105 |
| 0.0109 | 34.4086 | 3200 | 0.0002 | 0.3068 |
| 0.0106 | 36.5591 | 3400 | 0.0011 | 0.3081 |
| 0.0171 | 38.7097 | 3600 | 0.0012 | 0.3093 |
| 0.0099 | 40.8602 | 3800 | 0.0002 | 0.3130 |
| 0.0102 | 43.0108 | 4000 | 0.0014 | 0.3154 |
| 0.0129 | 45.1613 | 4200 | 0.0003 | 0.3105 |
| 0.0108 | 47.3118 | 4400 | 0.0001 | 0.3068 |
| 0.0085 | 49.4624 | 4600 | 0.0001 | 0.3093 |
| 0.0067 | 51.6129 | 4800 | 0.0001 | 0.3081 |
| 0.0079 | 53.7634 | 5000 | 0.0006 | 0.3068 |
| 0.0078 | 55.9140 | 5200 | 0.0001 | 0.3093 |
| 0.0091 | 58.0645 | 5400 | 0.0020 | 0.3081 |
| 0.0071 | 60.2151 | 5600 | 0.0017 | 0.3154 |
| 0.004 | 62.3656 | 5800 | 0.0001 | 0.3105 |
| 0.004 | 64.5161 | 6000 | 0.0001 | 0.3093 |
| 0.0064 | 66.6667 | 6200 | 0.0096 | 0.3166 |
| 0.0048 | 68.8172 | 6400 | 0.0000 | 0.3068 |
| 0.0037 | 70.9677 | 6600 | 0.0321 | 0.3081 |
| 0.0041 | 73.1183 | 6800 | 0.0000 | 0.3093 |
| 0.0059 | 75.2688 | 7000 | 0.0013 | 0.3093 |
| 0.0019 | 77.4194 | 7200 | 0.0011 | 0.3081 |
| 0.0022 | 79.5699 | 7400 | 0.0000 | 0.3068 |
| 0.0022 | 81.7204 | 7600 | 0.0000 | 0.3068 |
| 0.004 | 83.8710 | 7800 | 0.0000 | 0.3081 |
| 0.0025 | 86.0215 | 8000 | 0.0000 | 0.3081 |
| 0.0032 | 88.1720 | 8200 | 0.0000 | 0.3081 |
| 0.0019 | 90.3226 | 8400 | 0.0000 | 0.3093 |
| 0.001 | 92.4731 | 8600 | 0.0000 | 0.3081 |
| 0.001 | 94.6237 | 8800 | 0.0000 | 0.3093 |
| 0.0018 | 96.7742 | 9000 | 0.0000 | 0.3093 |
| 0.0019 | 98.9247 | 9200 | 0.0000 | 0.3093 |
### Framework versions
- Transformers 4.45.0.dev0
- Pytorch 2.4.0
- Datasets 2.21.0
- Tokenizers 0.19.1
|