Hey, sorry I wasn't able to replicate this error. Can you give me more details on the code?

Hi @Samyak29 Great work!
Can You share your inference script for me as a reference?

Sure, I am using this segmentation with whisperx library. Although I cannot share the whole script, I can share a part of it:
!pip install git+https://github.com/Hasan-Naseer/whisperX.git@release/latest-faster-whisper-version
!pip install ctranslate2==4.4.0
import whisperx
from diarizers import SegmentationModel

Load the fine-tuned segmentation model

segmentation_model = SegmentationModel().from_pretrained('Samyak29/speaker-segmentation-fine-tuned-hindi')

Convert to a Pyannote model

fine_tuned_model = segmentation_model.to_pyannote_model()

Replace the default segmentation model of whisperX/any other library you are using with the fine-tuned one

diarize_model.model._segmentation.model = fine_tuned_model.to(device)

diarize_segments = diarize_model(audio, num_speakers=2)

result = whisperx.assign_word_speakers(diarize_segments, result)