Recogment/titanet-large-onnx

TitaNet Large speaker embeddings (ONNX) — Recogment mirror

192-dim speaker-embedding model. This repo packages an ONNX export of the upstream NVIDIA NeMo checkpoint so that runtimes which don't depend on PyTorch / NeMo can use the model directly.

⚠️ Unofficial. This is a community-maintained mirror of derived artefacts. For the canonical model, citation, and training details see nvidia/speakerverification_en_titanet_large.

Files

File	Purpose	Size
titanet-large.onnx	Speaker embedding extractor (192-dim output)	~97 MB

Provenance

Exported from the upstream NeMo .nemo checkpoint of nvidia/speakerverification_en_titanet_large. No quantization, retraining, or fine-tuning — weights round-trip bit-for-bit through the export.

Why this mirror exists

The Recogment daemon ships with a pinned set of model SHA-256 hashes and a zero-egress sandbox; the only outbound network in the whole product is a companion downloader binary that fetches from public HTTPS sources. Until this mirror existed, TitaNet had to be shipped to beta testers out-of-band as a tarball. With this mirror in place, the downloader can fetch it automatically alongside the rest of the public catalogue.

If you're not building Recogment, you almost certainly want the official NVIDIA repo linked above instead — it includes the original .nemo checkpoint, configuration, and reference NeMo inference code.

License

This work is licensed under CC-BY-4.0, matching the upstream NVIDIA repo. Attribution: NVIDIA Corporation, via nvidia/speakerverification_en_titanet_large.

Citation

Please cite the upstream model paper, not this mirror:

@misc{nvidia2023titanet,
  title={TitaNet-Large: Speaker Verification with NeMo},
  author={NVIDIA},
  year={2023},
  publisher={Hugging Face},
  howpublished={\url{https://huggingface.co/nvidia/speakerverification_en_titanet_large}}
}