Downloads · 30 days
5
2% of all-time downloads
whisper-medium-afrispeech-all
robello2/whisper-medium-afrispeech-all
whisper-medium-afrispeech-all is a automatic speech recognition model from robello2. 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
214
Public
Parameters
764M
33.6 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors6.1 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Mar 11, 2026
- Updated
- Mar 15, 2026
- SHA
- a34c3c2c
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-medium-afrispeech-all
This model is a fine-tuned version of openai/whisper-medium on the afrispeech-200 dataset. It achieves the following results on the evaluation set:
- Loss: 0.7175
- Model Preparation Time: 0.0079
- Wer: 0.119
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: Use OptimizerNames.ADAMW_TORCH 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: 3589
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Model Preparation Time | Wer |
|---|---|---|---|---|---|
| 0.4502 | 1.0 | 3590 | 0.5078 | 0.0088 | 0.134 |
| 0.3523 | 2.0 | 7180 | 0.4575 | 0.0088 | 0.126 |
| 0.1643 | 3.0 | 10770 | 0.4776 | 0.0088 | 0.126 |
| 0.0719 | 4.0 | 14360 | 0.5220 | 0.0088 | 0.123 |
| 0.0402 | 5.0 | 17950 | 0.5615 | 0.0088 | 0.122 |
| 0.0157 | 6.0 | 21540 | 0.5964 | 0.0088 | 0.119 |
| 0.0088 | 7.0 | 25130 | 0.6425 | 0.0088 | 0.12 |
| 0.0042 | 8.0 | 28720 | 0.6639 | 0.0088 | 0.121 |
| 0.0007 | 9.0 | 32310 | 0.6972 | 0.0088 | 0.12 |
| 0.0005 | 10.0 | 35900 | 0.7175 | 0.0079 | 0.119 |
Framework versions
- Transformers 5.3.0
- Pytorch 2.6.0+cu124
- Datasets 2.19.0
- Tokenizers 0.22.2
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Created
- Mar 11, 2026
- Updated
- Mar 15, 2026