A study of semi-supervised speaker diarization system using gan mixture model
This work addresses speaker diarization for meeting analysis, presenting an incremental improvement by applying a recent clustering technique to an existing task.
The paper tackles speaker diarization by proposing a semi-supervised system using a generative adversarial network mixture model (GANMM) for clustering, achieving a best average diarization error rate of 17.11% on the AMI meeting corpus, which is a 33% relative improvement over an information bottleneck baseline.
We propose a new speaker diarization system based on a recently introduced unsupervised clustering technique namely, generative adversarial network mixture model (GANMM). The proposed system uses x-vectors as front-end representation. Spectral embedding is used for dimensionality reduction followed by k-means initialization during GANMM pre-training. GANMM performs unsupervised speaker clustering by efficiently capturing complex data distributions. Experimental results on the AMI meeting corpus show that the proposed semi-supervised diarization system matches or exceeds the performance of competitive baselines. On an evaluation set containing fifty sessions with varying durations, the best achieved average diarization error rate (DER) is 17.11%, a relative improvement of 33% over the information bottleneck baseline and comparable to xvector baseline.