Downloads · 30 days
0
tonythethompson/diar-streaming-sortformer-4spk-v2.1-onnx
diar-streaming-sortformer-4spk-v2.1-onnx is a automatic speech recognition model from tonythethompson. Use it when you need speech turned into text. The card lists the license as other.
This repository contains an ONNX export of NVIDIA's streaming Sortformer diarizer 4-speaker v2.1 model.
Downloads · 30 days
0
Access
Public
Updated Sep 30, 2026
Repo size
1 GB
Likes
0
Public
Click a slice to open those files.
.onnx755 MB · 74%
From the Hugging Face model README
This repository contains an ONNX export of NVIDIA's streaming Sortformer diarizer 4-speaker v2.1 model.
| Field | Value |
|---|---|
| Upstream model | nvidia/diar_streaming_sortformer_4spk-v2.1 |
| Upstream source revision | fafaab5faa1617a0ca52d38dd3dc4bd636800d3d |
| Export tool/script | ONNX export from NVIDIA Sortformer checkpoint (community packaging) |
| Quantization recipe | FP32 ONNX (onnx/model.onnx) |
Export tooling, precision, and quantization are recorded in the Source table above. This packaging mirror does not publish independent parity benchmarks; validate on your target execution provider before production use.
| File | Description |
|---|---|
onnx/model.onnx | ONNX model file |
Use the ONNX model for speaker diarization experiments or integration work where a streaming four-speaker diarization model is appropriate.
The upstream model card lists Fisher English, NIST SRE 2004-2010, LibriSpeech, AMI Meeting Corpus, VoxConverse v0.3, ICSI, AISHELL-4, DIHARD Challenge 3 development data, NIST SRE 2000 Disc8 split1, NOTSOFAR1, AliMeeting-train, and DiPCo.
The upstream model card reports Diarization Error Rate (DER) results across DIHARD III, CALLHOME, AliMeeting, AMI, NOTSOFAR1, and CH109 evaluation sets. This repository does not document ONNX-specific parity results.
Use of the source model is governed by the NVIDIA Open Model License Agreement.
trt_rtx_fp16/: mixed-precision FP16 candidate produced with NVIDIA ModelOpt.trt_rtx_fp16_graphsurgeon/: graph-cleaned copy using ONNX GraphSurgeon.The existing onnx/model.onnx remains unchanged. See verification_report.json for checks and limits. The candidates were exercised using ONNX Runtime's CUDA Execution Provider. TensorRT-RTX EP parsing and execution were not tested, so TensorRT-RTX compatibility is not claimed. Keep the .onnx.data sidecar adjacent to the GraphSurgeon graph.