Downloads · 30 days
2
5% of all-time downloads
wav2vec2-large-xls-r-300m-ko_Y2
Gummybear05/wav2vec2-large-xls-r-300m-ko_Y2
wav2vec2-large-xls-r-300m-ko_Y2 is a automatic speech recognition model from Gummybear05. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
39
Public
Parameters
317M
30.4 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1.3 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- wav2vec2
- Access
- Public
- Created
- Dec 20, 2023
- Updated
- Dec 20, 2023
- SHA
- 1c42e044
Base models
Model card
From the Hugging Face model README
<!-- 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. -->
wav2vec2-large-xls-r-300m-ko_Y2
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.7543
- Cer: 63.2284
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: 8
- 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: 100
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 30.4598 | 0.13 | 200 | 4.8628 | 100.0 |
| 4.9057 | 0.26 | 400 | 4.6582 | 100.0 |
| 4.8545 | 0.39 | 600 | 4.6698 | 100.0 |
| 4.8629 | 0.52 | 800 | 4.6900 | 100.0 |
| 4.8065 | 0.64 | 1000 | 4.6762 | 100.0 |
| 4.8111 | 0.77 | 1200 | 4.6486 | 100.0 |
| 4.7942 | 0.9 | 1400 | 4.8506 | 100.0 |
| 4.741 | 1.03 | 1600 | 4.6609 | 100.0 |
| 4.7087 | 1.16 | 1800 | 4.6844 | 100.0 |
| 4.6709 | 1.29 | 2000 | 4.5644 | 100.0 |
| 4.5943 | 1.42 | 2200 | 4.5284 | 100.0 |
| 4.4217 | 1.55 | 2400 | 4.3024 | 99.8297 |
| 3.9213 | 1.68 | 2600 | 3.9734 | 87.5822 |
| 3.4005 | 1.81 | 2800 | 3.5616 | 75.6931 |
| 2.9827 | 1.93 | 3000 | 3.4234 | 70.6473 |
| 2.7042 | 2.06 | 3200 | 3.3518 | 70.1304 |
| 2.5051 | 2.19 | 3400 | 3.0818 | 70.3947 |
| 2.3372 | 2.32 | 3600 | 2.8767 | 65.0082 |
| 2.2472 | 2.45 | 3800 | 2.9150 | 68.0804 |
| 2.1194 | 2.58 | 4000 | 2.8689 | 64.7380 |
| 2.0699 | 2.71 | 4200 | 2.8134 | 64.4149 |
| 1.9761 | 2.84 | 4400 | 2.8036 | 64.4796 |
| 1.9927 | 2.97 | 4600 | 2.7543 | 63.2284 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- License
- apache-2.0
- Created
- Dec 20, 2023
- Updated
- Dec 20, 2023