Improving Diarization Robustness using Diversification, Randomization and the DOVER Algorithm
This addresses robustness issues in speaker diarization for applications like meeting analysis, but it is incremental as it builds on existing methods like DOVER.
The paper tackled the problem of speaker diarization being sensitive to hyperparameters like cluster numbers and feature weighting, which are hard to optimize robustly across datasets, and showed that using the DOVER algorithm to average across parameter choices and diversify clustering outputs improved robustness and performance on conference meeting data sets.
Speaker diarization based on bottom-up clustering of speech segments by acoustic similarity is often highly sensitive to the choice of hyperparameters, such as the initial number of clusters and feature weighting. Optimizing these hyperparameters is difficult and often not robust across different data sets. We recently proposed the DOVER algorithm for combining multiple diarization hypotheses by voting. Here we propose to mitigate the robustness problem in diarization by using DOVER to average across different parameter choices. We also investigate the combination of diverse outputs obtained by following different merge choices pseudo-randomly in the course of clustering, thereby mitigating the greediness of best-first clustering. We show on two conference meeting data sets drawn from NIST evaluations that the proposed methods indeed yield more robust, and in several cases overall improved, results.