Downloads · 30 days
86
8% of all-time downloads
en_whisper_nonstandard_medium
Veronica1NW/en_whisper_nonstandard_medium
en_whisper_nonstandard_medium is a automatic speech recognition model from Veronica1NW. Use it when you need speech turned into text. The card lists the license as apache-2.0.
Languages: English (en) — Non-standard / accented varieties from Kenyan speakers
All-time downloads
1.1K
Public
Parameters
764M
3.1 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors3.1 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Oct 27, 2025
- Updated
- Oct 27, 2025
- SHA
- c4dffce3
Base models
Model card
From the Hugging Face model README
English Whisper Non-Standard Medium
Languages: English (en) — Non-standard / accented varieties from Kenyan speakers
This model was fine-tuned on accented and non-standard Kenyan English speech to enhance ASR accuracy for regional pronunciations.
Fine-tuning data: Subset of Mozilla Common Voice + curated Kenyan English speech
📊 Evaluation
| Metric | Dev | Test |
|---|---|---|
| WER | 0.1484 | 0.1107 |
| CER | 0.0810 | 0.0613 |
- Task
- Automatic Speech Recognition
- Type
- whisper
- License
- apache-2.0
- Languages
- en
- Created
- Oct 27, 2025
- Updated
- Oct 27, 2025