Downloads · 30 days
16
30% of all-time downloads
wav2vec2-nepali-commonvoice-ekg
RikeshSilwal/wav2vec2-nepali-commonvoice-ekg
wav2vec2-nepali-commonvoice-ekg is a automatic speech recognition model from RikeshSilwal. 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
53
Public
Repo size
22.7 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin1.3 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- wav2vec2
- Access
- Public
- Created
- Jan 2, 2024
- Updated
- Jan 2, 2024
- SHA
- 58e33307
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-nepali-commonvoice-ekg
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: 3.3075
- Wer: 0.9179
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: 3e-05
- train_batch_size: 14
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 28
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 60
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 22.4043 | 3.57 | 50 | 16.6522 | 1.0 |
| 10.7801 | 7.14 | 100 | 7.4333 | 1.0 |
| 6.495 | 10.71 | 150 | 5.7807 | 1.0 |
| 5.3133 | 14.29 | 200 | 4.8664 | 1.0 |
| 4.6032 | 17.86 | 250 | 4.2820 | 1.0 |
| 4.1089 | 21.43 | 300 | 3.9245 | 1.0 |
| 3.8458 | 25.0 | 350 | 3.7122 | 1.0 |
| 3.6591 | 28.57 | 400 | 3.5473 | 1.0 |
| 3.5109 | 32.14 | 450 | 3.4390 | 0.9415 |
| 3.4283 | 35.71 | 500 | 3.3844 | 0.9336 |
| 3.4034 | 39.29 | 550 | 3.3591 | 0.9858 |
| 3.3724 | 42.86 | 600 | 3.3384 | 0.9842 |
| 3.3415 | 46.43 | 650 | 3.3256 | 0.9795 |
| 3.3093 | 50.0 | 700 | 3.3183 | 0.9668 |
| 3.2933 | 53.57 | 750 | 3.3106 | 0.9384 |
| 3.2663 | 57.14 | 800 | 3.3075 | 0.9179 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.13.3
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- License
- apache-2.0
- Created
- Jan 2, 2024
- Updated
- Jan 2, 2024