diar_streaming_sortformer_4spk-v2

Home/Models/nvidia/diar_streaming_sortformer_4spk-v2

nvidia/diar_streaming_sortformer_4spk-v2

kind: original
quant: bf16 (16-bit)
family: misc
modality: automatic-speech-recognition
root model: nvidia/diar_streaming_sortformer_4spk-v2
provider: nvidia
license: cc-by-4.0
language: —
context: —
library: nemo
vRAM (GB): —
size on disk: —
downloads: 43,723
likes: 131
created: —
last modified: 2025-12-31T23:04:03+00:00 (2025)
huggingface: https://huggingface.co/nvidia/diar_streaming_sortformer_4spk-v2
tags: nemospeaker-diarizationspeaker-recognitionspeechaudioTransformerFastConformerConformerNESTpytorchNeMoautomatic-speech-recognitiondataset:fisher_englishdataset:NIST_SRE_2004-2010dataset:librispeechdataset:ami_meeting_corpusdataset:voxconverse_v0.3dataset:icsidataset:aishell4dataset:dihard_challenge-3-devdataset:NIST_SRE_2000-Disc8_split1dataset:Alimeeting-traindataset:DiPCoarxiv:2409.06656

benchmark	category	score	context	conf	evidence
RTF	speech-asr	0.2		low	`\| very high latency \| 30.4s \| 0.002 \| 340 \| 40 \| 40 \| 300 \| 188 \|`

none scraped.