Downloads · 30 days
0
godai-african-asr
techsalomon/godai-african-asr
godai-african-asr is a automatic speech recognition model from techsalomon. Use it when you need speech turned into text. The card lists the license as cc-by-nc-4.0.
- adapterfonbest.pt — best WER across ALL sessions (use in production) - adapterfonlatest.pt — most recent checkpoint - vocabfon.json — vocabulary - werlogfon.txt — full training history
Access
Public
Updated Aug 30, 2026
Repo size
2 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt109 MB · 100%
At a glance
- Task
- Automatic Speech Recognition
- License
- cc-by-nc-4.0
- Access
- Public
- Created
- Jun 28, 2026
- Updated
- Aug 30, 2026
- SHA
- b13cb30f
Base models
Model card
From the Hugging Face model README
God AI 🇧🇯 — Fon Cotonou ASR Adapter
| Property | Value |
|---|---|
| Base model | facebook/mms-1b-all |
| Dialect | Fon Cotonou |
| Current step | 1250 |
| Current WER | 0.826 |
| Current CER | 0.340 |
| Best WER ever | 0.820 |
| Platform | https://www.godai.africa |
| Voice API | https://voice.godai.africa/transcribe |
| Research | https://doi.org/10.5281/zenodo.21348209 |
| Founder | Djokoto Edem Salomon — Tech Salomon, Lomé, Togo |
Files
- adapter_fon_best.pt — best WER across ALL sessions (use in production)
- adapter_fon_latest.pt — most recent checkpoint
- vocab_fon.json — vocabulary
- wer_log_fon.txt — full training history
- Task
- Automatic Speech Recognition
- License
- cc-by-nc-4.0
- Created
- Jun 28, 2026
- Updated
- Aug 30, 2026