Downloads · 30 days
0
baharani_v12_lora
9FA/baharani_v12_lora
baharani_v12_lora is a machine learning model from 9FA. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
LoRA fine-tune of openbmb/VoxCPM2 on wldaldakheel/baharani-mix.
Access
Public
Updated Aug 23, 2026
Repo size
582 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pth290 MB · 50%
At a glance
- Access
- Public
- Created
- Aug 23, 2026
- Updated
- Aug 23, 2026
- SHA
- de8b27f1
Model card
From the Hugging Face model README
Baharani v12 VoxCPM2 LoRA
LoRA fine-tune of openbmb/VoxCPM2 on wldaldakheel/baharani-mix.
v12 Training Recipe
| Param | Value |
|---|---|
| LoRA rank | 64 |
| LoRA alpha | 128 (ratio 2.0) |
| LoRA dropout | 0.0 (implicit regularization from rank constraint) |
| Learning rate | 5e-5 |
| Scheduler | Cosine with 1000-step warmup |
| Steps | 15000 (~3.1 epochs) |
| Batch size | 4 × 4 grad_accum = 16 effective |
| EMA decay | 0.999 |
| Validation | Every 3000 steps, best checkpoint selected |
Key improvements over v10
- LoRA rank 64/alpha=128 (was r=32/alpha=32): more capacity for dialect features
- Dropout removed: 68h of data provides natural regularization; dropout hurts audio quality
- 3.1 epochs (was ~1 epoch): proper convergence without underfitting
- Cosine LR: smooth decay instead of constant → better convergence
- EMA decay=0.999: appropriate halflife for 15K training steps
- Train/val split: best checkpoint selection by validation loss
- Lower LR (5e-5 vs 1e-4): finer adaptation without catastrophic forgetting
Usage
For EMA weights (recommended for production):
License
Apache 2.0
- Created
- Aug 23, 2026
- Updated
- Aug 23, 2026