diar_streaming_sortformer_4spk-v2.1-fp32
mlx-community/diar_streaming_sortformer_4spk-v2.1-fp32
Voice activity detection · mlx-audio
fp32 Open on Hugging Face
Parameters
118M
Weights (est.)
≈ 472 MB
Mac memory (est.)
8 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Details
- Base model
- nvidia/diar_streaming_sortformer_4spk-v2.1 (finetune)
- Vendor
- NVIDIA
- Task
- Voice activity detection
- Quantisation
- fp32
- Licence
- —
- Downloads
- 226 in the last 30 days · 1.1K all time · 2 likes
- Published
- 2026-02-10 · updated 2026-02-10 (9mo ago)
- Tags
- mlx-audio, sortformer, speaker-diarization, speech, streaming
Other MLX builds of nvidia/diar_streaming_sortformer_4spk-v2.1
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| diar_streaming_sortformer_4spk-v2.1-fp16 | fp16 | ≈ 236 MB | 8 GB+ | 1.1K |
| diar_streaming_sortformer_4spk-v2.1-fp32 (this one) | fp32 | ≈ 472 MB | 8 GB+ | 226 |