Downloads · 30 days
13
32% of all-time downloads
wav2vec2-base-timit-demo-google-colab
hirohiroz/wav2vec2-base-timit-demo-google-colab
wav2vec2-base-timit-demo-google-colab is a automatic speech recognition model from hirohiroz. 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
41
Public
Repo size
755 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin378 MB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- wav2vec2
- Access
- Public
- Created
- Jul 10, 2022
- Updated
- Jul 10, 2022
- SHA
- 1945f2ae
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-base-timit-demo-google-colab
This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5173
- Wer: 0.3399
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: 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: 1000
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 3.5684 | 1.0 | 500 | 2.1662 | 1.0068 |
| 0.9143 | 2.01 | 1000 | 0.5820 | 0.5399 |
| 0.439 | 3.01 | 1500 | 0.4596 | 0.4586 |
| 0.3122 | 4.02 | 2000 | 0.4623 | 0.4181 |
| 0.2391 | 5.02 | 2500 | 0.4243 | 0.3938 |
| 0.1977 | 6.02 | 3000 | 0.4421 | 0.3964 |
| 0.1635 | 7.03 | 3500 | 0.5076 | 0.3977 |
| 0.145 | 8.03 | 4000 | 0.4639 | 0.3754 |
| 0.1315 | 9.04 | 4500 | 0.5181 | 0.3652 |
| 0.1131 | 10.04 | 5000 | 0.4496 | 0.3778 |
| 0.1005 | 11.04 | 5500 | 0.4438 | 0.3664 |
| 0.0919 | 12.05 | 6000 | 0.4868 | 0.3865 |
| 0.0934 | 13.05 | 6500 | 0.5163 | 0.3694 |
| 0.076 | 14.06 | 7000 | 0.4543 | 0.3719 |
| 0.0727 | 15.06 | 7500 | 0.5296 | 0.3807 |
| 0.0657 | 16.06 | 8000 | 0.4715 | 0.3699 |
| 0.0578 | 17.07 | 8500 | 0.4927 | 0.3699 |
| 0.057 | 18.07 | 9000 | 0.4767 | 0.3660 |
| 0.0493 | 19.08 | 9500 | 0.5306 | 0.3623 |
| 0.0425 | 20.08 | 10000 | 0.4828 | 0.3561 |
| 0.0431 | 21.08 | 10500 | 0.4875 | 0.3620 |
| 0.0366 | 22.09 | 11000 | 0.4984 | 0.3482 |
| 0.0332 | 23.09 | 11500 | 0.5375 | 0.3477 |
| 0.0348 | 24.1 | 12000 | 0.5406 | 0.3361 |
| 0.0301 | 25.1 | 12500 | 0.4954 | 0.3381 |
| 0.0294 | 26.1 | 13000 | 0.5033 | 0.3424 |
| 0.026 | 27.11 | 13500 | 0.5254 | 0.3384 |
| 0.0243 | 28.11 | 14000 | 0.5189 | 0.3402 |
| 0.0221 | 29.12 | 14500 | 0.5173 | 0.3399 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.11.0+cu113
- Datasets 1.18.3
- Tokenizers 0.12.1
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- License
- apache-2.0
- Created
- Jul 10, 2022
- Updated
- Jul 10, 2022