how to run the pipeline with different window and step size ?

[amitli1]

I want to run the pipeline with a window size of 3 seconds and 1-second step.

from pyannote.audio import Pipeline
pipeline = Pipeline.from_pretrained("pyannote/speaker-diarization",use_auth_token="..")
diarization = pipeline("example.wav")

I didn't see any configuration I can pass to the pipeline.

How can I do it ?

[hbredin]

This is not supported out of the box.
You'd have to modify the code and pass your required duration and step here:

pyannote-audio/pyannote/audio/pipelines/speaker_diarization.py

Line 141 in 74939ac

self._segmentation = Inference(

[Rajsharma82]

File "C:\spreco\pyno.py", line 6, in
diarization = pipeline("eng.wav")
^^^^^^^^^^^^^^^^^^^
TypeError: 'NoneType' object is not callable

when i am trying to run this code this error is occuring