Downloads · 30 days
6
8% of all-time downloads
whisper-small-af-ZA
UnaiGurbindo/whisper-small-af-ZA
whisper-small-af-ZA is a automatic speech recognition model from UnaiGurbindo. 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
76
Public
Parameters
242M
21.3 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors967 MB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Feb 4, 2024
- Updated
- Feb 5, 2024
- SHA
- e20e20a3
Base models
Datasets
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-af-ZA
This model is a fine-tuned version of openai/whisper-tiny on the PolyAI/minds14 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0415
- Wer Ortho: 0.0529
- Wer: 0.0293
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 5
- training_steps: 700
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 0.0054 | 1.45 | 100 | 0.0312 | 0.0449 | 0.0228 |
| 0.0025 | 2.9 | 200 | 0.0345 | 0.0456 | 0.0231 |
| 0.0021 | 4.35 | 300 | 0.0325 | 0.0445 | 0.0206 |
| 0.0018 | 5.8 | 400 | 0.0325 | 0.0449 | 0.0202 |
| 0.0033 | 7.25 | 500 | 0.0390 | 0.0905 | 0.0654 |
| 0.0043 | 8.7 | 600 | 0.0415 | 0.0577 | 0.0347 |
| 0.0026 | 10.14 | 700 | 0.0415 | 0.0529 | 0.0293 |
Framework versions
- Transformers 4.38.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Created
- Feb 4, 2024
- Updated
- Feb 5, 2024