Downloads · 30 days
17
3% of all-time downloads
whisper-lv-small-augmented
ALM/whisper-lv-small-augmented
whisper-lv-small-augmented is a automatic speech recognition model from ALM. 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
528
Public
Parameters
242M
5.8 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.bin967 MB · 50%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Dec 21, 2022
- Updated
- Aug 28, 2023
- SHA
- 7a673b6b
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. -->
Whisper Small Latvian - Robust
This model is a fine-tuned version of openai/whisper-small on the mozilla-foundation/common_voice_11_0 lv dataset. It achieves the following results on the evaluation set:
- Loss: 0.5621
- Wer: 33.1120
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: 5e-05
- train_batch_size: 64
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0219 | 16.67 | 1000 | 0.6129 | 41.1575 |
| 0.0029 | 33.33 | 2000 | 0.5975 | 36.1480 |
| 0.0003 | 50.0 | 3000 | 0.5626 | 33.6812 |
| 0.0004 | 66.67 | 4000 | 0.5608 | 33.2827 |
| 0.0001 | 83.33 | 5000 | 0.5715 | 33.8046 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.8.0
- Tokenizers 0.13.2
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Languages
- lv
- Created
- Dec 21, 2022
- Updated
- Aug 28, 2023