ISCA Archive Interspeech 2021
ISCA Archive Interspeech 2021

Speaker-Conversation Factorial Designs for Diarization Error Analysis

Scott Seyfarth, Sundararajan Srinivasan, Katrin Kirchhoff

Speaker diarization accuracy can be affected by both acoustics and conversation characteristics. Determining the cause of diarization errors is difficult because speaker voice acoustics and conversation structure co-vary, and the interactions between acoustics, conversational structure, and diarization accuracy are complex. This paper proposes a methodology that can distinguish independent marginal effects of acoustic and conversation characteristics on diarization accuracy by remixing conversations in a factorial design. As an illustration, this approach is used to investigate gender-related and language-related accuracy differences with three diarization systems: a baseline system using subsegment x-vector clustering, a variant of it with shorter subsegments, and a third system based on a Bayesian hidden Markov model. Our analysis shows large accuracy disparities for the baseline system primarily due to conversational structure, which are partially mitigated in the other two systems. The illustration thus demonstrates how the methodology can be used to identify and guide diarization model improvements.

doi: 10.21437/Interspeech.2021-1864

Cite as: Seyfarth, S., Srinivasan, S., Kirchhoff, K. (2021) Speaker-Conversation Factorial Designs for Diarization Error Analysis. Proc. Interspeech 2021, 756-760, doi: 10.21437/Interspeech.2021-1864

  author={Scott Seyfarth and Sundararajan Srinivasan and Katrin Kirchhoff},
  title={{Speaker-Conversation Factorial Designs for Diarization Error Analysis}},
  booktitle={Proc. Interspeech 2021},