EngineX-Iluvatar/enginex-mr_series-sherpa-onnx
Archived
16
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

Export speaker verification models from NeMo to ONNX (#526)

Browse Source
This commit is contained in:
Fangjun Kuang
2024-01-13 19:49:45 +08:00
committed by GitHub
parent afc81ec122
commit 68a525a024
10 changed files with 448 additions and 28 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
scripts/nemo/speaker-verification/README.md Normal file
View File

@@ -0,0 +1,14 @@
# Introduction
This directory contains script for exporting speaker verification models
from [NeMo](https://github.com/NVIDIA/NeMo/) to onnx
so that you can use them in `sherpa-onnx`.
Specifically, the following 4 models are exported to `sherpa-onnx`
from
[this page](https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/main/asr/speaker_recognition/results.html#speaker-recognition-models):
- [titanet_large](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/nemo/models/titanet_large),
- [titanet_small](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/nemo/models/titanet_small)
- [speakerverification_speakernet](https://ngc.nvidia.com/catalog/models/nvidia:nemo:speakerverification_speakernet)
- [ecapa_tdnn](https://ngc.nvidia.com/catalog/models/nvidia:nemo:ecapa_tdnn)