Downloads · 30 days
13
37% of all-time downloads
wav2vec2-demo-F03
yip-i/wav2vec2-demo-F03
wav2vec2-demo-F03 is a automatic speech recognition model from yip-i. Use it when you need speech turned into text. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
35
Public
Repo size
3 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt1.4 GB · 66%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- Model type
- wav2vec2
- Access
- Public
- Created
- Nov 15, 2022
- Updated
- Nov 20, 2022
- SHA
- 9231aa7d
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-demo-F03
This model is a fine-tuned version of yip-i/uaspeech-pretrained on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.8742
- Wer: 1.2914
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
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 6.4808 | 0.97 | 500 | 3.0628 | 1.1656 |
| 2.9947 | 1.94 | 1000 | 3.0334 | 1.1523 |
| 2.934 | 2.91 | 1500 | 3.0520 | 1.1648 |
| 2.9317 | 3.88 | 2000 | 3.3808 | 1.0 |
| 3.0008 | 4.85 | 2500 | 3.0342 | 1.2559 |
| 3.112 | 5.83 | 3000 | 3.1228 | 1.1258 |
| 2.8972 | 6.8 | 3500 | 2.9885 | 1.2914 |
| 2.8911 | 7.77 | 4000 | 3.2586 | 1.2754 |
| 2.9884 | 8.74 | 4500 | 3.0487 | 1.2090 |
| 2.873 | 9.71 | 5000 | 2.9382 | 1.2914 |
| 3.3551 | 10.68 | 5500 | 3.2607 | 1.2844 |
| 3.6426 | 11.65 | 6000 | 3.0053 | 1.0242 |
| 2.9184 | 12.62 | 6500 | 2.9219 | 1.2828 |
| 2.8384 | 13.59 | 7000 | 2.9530 | 1.2816 |
| 2.8855 | 14.56 | 7500 | 2.9978 | 1.0121 |
| 2.8479 | 15.53 | 8000 | 2.9722 | 1.0977 |
| 2.8241 | 16.5 | 8500 | 2.9670 | 1.3082 |
| 2.807 | 17.48 | 9000 | 2.9841 | 1.2914 |
| 2.8115 | 18.45 | 9500 | 2.9484 | 1.2977 |
| 2.8123 | 19.42 | 10000 | 2.9310 | 1.2914 |
| 3.0291 | 20.39 | 10500 | 2.9665 | 1.2902 |
| 2.8735 | 21.36 | 11000 | 2.9245 | 1.1160 |
| 2.8164 | 22.33 | 11500 | 2.9137 | 1.2914 |
| 2.8084 | 23.3 | 12000 | 2.9543 | 1.1891 |
| 2.8079 | 24.27 | 12500 | 2.9179 | 1.4516 |
| 2.7916 | 25.24 | 13000 | 2.8971 | 1.2926 |
| 2.7824 | 26.21 | 13500 | 2.8990 | 1.2914 |
| 2.7555 | 27.18 | 14000 | 2.9004 | 1.2914 |
| 2.7803 | 28.16 | 14500 | 2.8747 | 1.2910 |
| 2.753 | 29.13 | 15000 | 2.8742 | 1.2914 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu113
- Datasets 1.18.3
- Tokenizers 0.13.2
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- Created
- Nov 15, 2022
- Updated
- Nov 20, 2022