Downloads · 30 days
2
10% of all-time downloads
length_6_sd
Theju/length_6_sd
length_6_sd is a automatic speech recognition model from Theju. 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
20
Public
Repo size
2.5 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
- Feb 24, 2023
- Updated
- Feb 24, 2023
- SHA
- 6099bfaf
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. -->
length_6_sd
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: 0.0702
- Wer: 1.6075
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: 4
- 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: 7
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 9.7985 | 0.68 | 500 | 4.6059 | 1.0 |
| 3.0404 | 1.36 | 1000 | 2.3474 | 2.4424 |
| 1.7444 | 2.04 | 1500 | 0.8992 | 2.3925 |
| 0.7667 | 2.72 | 2000 | 0.2616 | 1.8754 |
| 0.36 | 3.41 | 2500 | 0.1904 | 1.8598 |
| 0.2519 | 4.09 | 3000 | 0.0765 | 1.7290 |
| 0.152 | 4.77 | 3500 | 0.0902 | 1.5296 |
| 0.1052 | 5.45 | 4000 | 0.0842 | 1.5545 |
| 0.0773 | 6.13 | 4500 | 0.0660 | 1.6324 |
| 0.0729 | 6.81 | 5000 | 0.0702 | 1.6075 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.13.1+cu116
- Datasets 1.18.3
- Tokenizers 0.13.2
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- License
- apache-2.0
- Created
- Feb 24, 2023
- Updated
- Feb 24, 2023