Downloads · 30 days
8
22% of all-time downloads
pyannote-segmentation-bengali-finetuned
smam/pyannote-segmentation-bengali-finetuned
pyannote-segmentation-bengali-finetuned is a machine learning model from smam. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Fine-tuned pyannote/segmentation-3.0 for Bengali speaker diarization.
All-time downloads
37
Public
Repo size
23.6 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.ckpt17.7 MB · 75%
At a glance
- Access
- Public
- Created
- Feb 22, 2026
- Updated
- Feb 22, 2026
- SHA
- 2cf09402
Model card
From the Hugging Face model README
Bengali Speaker Diarization - Fine-tuned Segmentation Model
Fine-tuned pyannote/segmentation-3.0 for Bengali speaker diarization.
Results
- Best DER: 0.1312
Training Data (~370 hours)
- Synthetic V4: 600 files (~300 hours) from smam/bengali-diarization-synthetic-v4
- DISPLACE24: 67 files (~35 hours)
- DISPLACE26: 125 files (~35 hours)
Training
- Train samples: 600
- Val samples: 67
- Max epochs: 30 (with early stopping on DER, patience=5)
- Optimizer: AdamW (lr=0.0001, weight_decay=0.01)
- Precision: 16-mixed on H100
Task Configuration
- Duration: 10.0s per chunk
- Max speakers per chunk: 3
- Max speakers per frame: 2 (powerset encoding with overlap support)
Usage
from pyannote.audio import Model
import torch
# Load weights
state_dict = torch.load('pytorch_model.bin')
model = Model.from_pretrained('pyannote/segmentation-3.0')
model.load_state_dict(state_dict)
- Created
- Feb 22, 2026
- Updated
- Feb 22, 2026