Downloads · 30 days
5
22% of all-time downloads
wav2vec2-1b-Y
Gummybear05/wav2vec2-1b-Y
wav2vec2-1b-Y 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
23
Public
Parameters
964M
73.3 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors3.9 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- wav2vec2
- Access
- Public
- Created
- Nov 25, 2024
- Updated
- Nov 27, 2024
- SHA
- 0905e959
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. -->
wav2vec2-1b-Y
This model is a fine-tuned version of facebook/wav2vec2-xls-r-1b on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9820
- Cer: 24.1248
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: 1
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 16
- 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: 50
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 9.8741 | 0.2580 | 200 | 2.9382 | 58.9579 |
| 1.6995 | 0.5161 | 400 | 2.0333 | 45.6238 |
| 1.151 | 0.7741 | 600 | 1.7032 | 39.6734 |
| 0.9663 | 1.0322 | 800 | 1.2993 | 31.3675 |
| 0.8027 | 1.2902 | 1000 | 1.2846 | 33.0768 |
| 0.7227 | 1.5483 | 1200 | 1.1823 | 28.6419 |
| 0.6516 | 1.8063 | 1400 | 1.2823 | 32.2838 |
| 0.6087 | 2.0643 | 1600 | 1.2643 | 31.0209 |
| 0.5242 | 2.3224 | 1800 | 1.2452 | 30.3865 |
| 0.4763 | 2.5804 | 2000 | 1.1365 | 28.0839 |
| 0.4611 | 2.8385 | 2200 | 1.0796 | 26.5742 |
| 0.4103 | 3.0965 | 2400 | 1.1832 | 29.3527 |
| 0.3289 | 3.3546 | 2600 | 1.0230 | 25.0705 |
| 0.31 | 3.6126 | 2800 | 0.9800 | 24.8708 |
| 0.2995 | 3.8707 | 3000 | 0.9924 | 25.1880 |
| 0.2516 | 4.1287 | 3200 | 1.0370 | 25.2173 |
| 0.227 | 4.3867 | 3400 | 1.0256 | 24.8531 |
| 0.2251 | 4.6448 | 3600 | 0.9982 | 24.1424 |
| 0.2251 | 4.9028 | 3800 | 0.9820 | 24.1248 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.3.1.post100
- Datasets 2.19.1
- Tokenizers 0.20.1
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- License
- apache-2.0
- Created
- Nov 25, 2024
- Updated
- Nov 27, 2024