Systematic Evaluation Of Online Speaker Diarization Systems Regarding Their Latency

Abstract

In this paper, different online speaker diarization systems are evaluated on the same hardware with the same test data with regard to their latency. The latency is the time span from audio input to the output of the corresponding speaker label. As part of the evaluation, various model combinations within the DIART framework, a diarization system based on the online clustering algorithm UIS-RNN-SML, and the end-to-end online diarization system FS-EEND are compared. The lowest latency is achieved for the DIART-pipeline with the embedding model pyannote/embedding and the segmentation model pyannote/segmentation. The FS-EEND system shows a similarly good latency. In general there is currently no published research that compares several online diarization systems in terms of their latency. This makes this work even more relevant.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyaperdannier2024systematic

Related papers

• Overlap-aware Low-latency Online Speaker Diarization Based On End-to-end Local Segmentation (2021)10.35
• An Approach To Optimize Inference Of The DIART Speaker Diarization Pipeline (2024)0.00
• Low-latency Online Speaker Diarization With Graph-based Label Generation (2021)8.60
• An Experimental Review Of Speaker Diarization Methods With Application To Two-speaker Conversational Telephone Speech Recordings (2023)8.35
• Enhancements For Audio-only Diarization Systems (2019)0.00
• A Reinforcement Learning Framework For Online Speaker Diarization (2023)0.00
• Speaker Diarization With LSTM (2017)17.48
• Low-latency Speech Separation Guided Diarization For Telephone Conversations (2022)6.77