metadata
license: apache-2.0
base_model: facebook/wav2vec2-xls-r-300m
tags:
- generated_from_trainer
model-index:
- name: wav2vec2-large-xls-r-korean
results: []
wav2vec2-large-xls-r-korean
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5006
- Cer: 0.1178
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: 0.0001
- train_batch_size: 16
- 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: 1000
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 3.3321 | 1.11 | 300 | 3.8704 | 1.0 |
| 1.6999 | 2.22 | 600 | 1.8098 | 0.3638 |
| 0.6728 | 3.33 | 900 | 0.7710 | 0.1967 |
| 0.4398 | 4.44 | 1200 | 0.5838 | 0.1568 |
| 0.3202 | 5.56 | 1500 | 0.5421 | 0.1468 |
| 0.2286 | 6.67 | 1800 | 0.4912 | 0.1312 |
| 0.161 | 7.78 | 2100 | 0.4931 | 0.1247 |
| 0.1436 | 8.89 | 2400 | 0.4953 | 0.1211 |
| 0.1213 | 10.0 | 2700 | 0.5006 | 0.1178 |
Framework versions
- Transformers 4.33.2
- Pytorch 1.12.1+cu113
- Datasets 2.14.5
- Tokenizers 0.13.3