Downloads · 30 days
18
49% of all-time downloads
wav2vec2-1b-E30_freq_speed
Gummybear05/wav2vec2-1b-E30_freq_speed
wav2vec2-1b-E30_freq_speed 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
37
Public
Parameters
964M
42.4 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
- Dec 1, 2024
- Updated
- Dec 1, 2024
- SHA
- d35958a0
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-E30_freq_speed
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: 0.6382
- Cer: 17.3637
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: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- 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 |
|---|---|---|---|---|
| 29.1406 | 0.2580 | 200 | 16.8395 | 83.7289 |
| 10.9521 | 0.5160 | 400 | 4.7348 | 94.3550 |
| 9.351 | 0.7741 | 600 | 4.2935 | 92.1346 |
| 7.4517 | 1.0321 | 800 | 4.0701 | 79.2704 |
| 5.4533 | 1.2901 | 1000 | 3.5394 | 66.4767 |
| 3.4356 | 1.5481 | 1200 | 1.6827 | 37.0301 |
| 1.4001 | 1.8062 | 1400 | 1.3482 | 32.4953 |
| 1.063 | 2.0642 | 1600 | 1.1849 | 32.9359 |
| 0.8329 | 2.3222 | 1800 | 1.3395 | 32.4307 |
| 0.7532 | 2.5802 | 2000 | 1.1368 | 29.5700 |
| 0.6628 | 2.8383 | 2200 | 1.0047 | 28.0545 |
| 0.564 | 3.0963 | 2400 | 0.8462 | 24.4831 |
| 0.4531 | 3.3543 | 2600 | 0.7833 | 19.9366 |
| 0.3946 | 3.6123 | 2800 | 0.7219 | 18.7383 |
| 0.3829 | 3.8703 | 3000 | 0.7232 | 19.0848 |
| 0.3109 | 4.1284 | 3200 | 0.7258 | 19.6370 |
| 0.2561 | 4.3864 | 3400 | 0.6731 | 18.6736 |
| 0.2392 | 4.6444 | 3600 | 0.6570 | 17.9629 |
| 0.228 | 4.9024 | 3800 | 0.6382 | 17.3637 |
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
- Dec 1, 2024
- Updated
- Dec 1, 2024