Auto-Tuning Spectral Clustering for Speaker Diarization Using Normalized Maximum Eigengap

Abstract

In this study, we propose a new spectral clustering framework that can auto-tune the parameters of the clustering algorithm in the context of speaker diarization. The proposed framework uses normalized maximum eigengap (NME) values to estimate the number of clusters and the parameters for the threshold of the elements of each row in an affinity matrix during spectral clustering, without the use of parameter tuning on the development set. Even through this hands-off approach, we achieve a comparable or better performance across various evaluation sets than the results found using traditional clustering methods that apply careful parameter tuning and development data. A relative improvement of 17% in the speaker error rate on the well-known CALLHOME evaluation set shows the effectiveness of our proposed spectral clustering with auto-tuning.

in IEEE Signal Processing Letters, 2020

References in corpus (1)

• LSTM based Similarity Measurement with Spectral Clustering for Speaker Diarization

Cited by in corpus (9)

• Encoder-Decoder Based Attractors for End-to-End Neural Diarization
• TS-SEP: Joint Diarization and Separation Conditioned on Estimated Speaker Embeddings
• Online Neural Diarization of Unlimited Numbers of Speakers Using Global and Local Attractors
• Self-supervised Representation Learning With Path Integral Clustering For Speaker Diarization
• TSUP Speaker Diarization System for Conversational Short-phrase Speaker Diarization Challenge
• Community Detection Graph Convolutional Network for Overlap-Aware Speaker Diarization
• End-to-End Integration of Speech Separation and Voice Activity Detection for Low-Latency Diarization of Telephone Conversations
• Numerical Performance of the Implicitly Restarted Arnoldi Method in OFP8, Bfloat16, Posit, and Takum Arithmetics
• Self-Tuning Spectral Clustering for Speaker Diarization