How to inference with pretrained models

wenet-e2e / wespeaker

Research and Production Oriented Speaker Verification, Recognition and Diarization Toolkit

Apache License 2.0

738 stars 123 forks source link

How to inference with pretrained models #360

Open freshpearYoon opened 2 months ago

freshpearYoon commented 2 months ago

How can I perform SV with other pretrained models on my data ? Can I get example codes?

JiJiJiang commented 2 months ago

Yes, prepare your own data.scp and utt2spk and trials. You can follow examples/v2/run.sh stage 4 and stage 5, as well as data preparing

freshpearYoon commented 2 months ago

Is it impossible to inference on my single data with python file?

JiJiJiang commented 2 months ago

Of course you can, check cli usage.

Jiltseb commented 1 month ago

I was trying to use vblinkf model (SimAMResNet100) but gets an input mismatch error compared to SimAMResNet34.

nnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Got invalid dimensions for input: feats for the following indices
 index: 1 Got: 23 Expected: 200
 Please fix either the inputs/outputs or the model.

Greatly appreciate the help @JiJiJiang