Downloads · 30 days
5
2% of all-time downloads
nyanja-asr-whisper-tiny
buumba641/nyanja-asr-whisper-tiny
nyanja-asr-whisper-tiny is a automatic speech recognition model from buumba641. 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
255
Public
Parameters
37.8M
2.4 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors151 MB · 97%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- Model type
- whisper
- Access
- Public
- Created
- Apr 4, 2026
- Updated
- May 22, 2026
- SHA
- c23e72d6
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. -->
nyanja-asr-whisper-tiny
This model is a fine-tuned version of buumba641/nyanja-asr-whisper-tiny on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.0349
- Wer: 0.9721
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: 2e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.8296 | 2.7027 | 100 | 1.8980 | 1.2775 |
| 1.4551 | 5.4054 | 200 | 1.8875 | 1.0809 |
| 0.6801 | 8.1081 | 300 | 1.9566 | 1.0432 |
| 0.4784 | 10.8108 | 400 | 2.0349 | 0.9721 |
| 0.2899 | 13.5135 | 500 | 2.1027 | 0.9847 |
| 0.1860 | 16.2162 | 600 | 2.1898 | 0.9777 |
| 0.1292 | 18.9189 | 700 | 2.2167 | 1.0126 |
| 0.0690 | 21.6216 | 800 | 2.2689 | 1.0558 |
| 0.0401 | 24.3243 | 900 | 2.2870 | 1.0098 |
| 0.0350 | 27.0270 | 1000 | 2.3181 | 1.0349 |
| 0.0183 | 29.7297 | 1100 | 2.3307 | 1.2706 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- Created
- Apr 4, 2026
- Updated
- May 22, 2026