Target-Speaker Voice Activity Detection: a Novel Approach for Multi-Speaker Diarization in a Dinner Party Scenario

• Published in: arXiv.org 2020-07
• Main Authors: Medennikov, Ivan, Korenevsky, Maxim, Prisyach, Tatiana, Khokhlov, Yuri, Korenevskaya, Mariya, Sorokin, Ivan, Timofeeva, Tatiana, Mitrofanov, Anton, Andrusenko, Andrei, Podluzhny, Ivan, Laptev, Aleksandr, Romanenko, Aleksei
• Format: Article
• Language: English
• Subjects: Clustering; Post-processing; Target detection; Voice activity detectors; Voice recognition
• ISSN: 2331-8422
• DOI: 10.48550/arxiv.2005.07272

Speaker diarization for real-life scenarios is an extremely challenging problem. Widely used clustering-based diarization approaches perform rather poorly in such conditions, mainly due to the limited ability to handle overlapping speech. We propose a novel Target-Speaker Voice Activity Detection (TS-VAD) approach, which directly predicts an activity of each speaker on each time frame. TS-VAD model takes conventional speech features (e.g., MFCC) along with i-vectors for each speaker as inputs. A set of binary classification output layers produces activities of each speaker. I-vectors can be estimated iteratively, starting with a strong clustering-based diarization. We also extend the TS-VAD approach to the multi-microphone case using a simple attention mechanism on top of hidden representations extracted from the single-channel TS-VAD model. Moreover, post-processing strategies for the predicted speaker activity probabilities are investigated. Experiments on the CHiME-6 unsegmented data show that TS-VAD achieves state-of-the-art results outperforming the baseline x-vector-based system by more than 30% Diarization Error Rate (DER) abs.