A Teacher-Student approach for extracting informative speaker embeddings from speech mixtures

Abstract

We introduce a monaural neural speaker embeddings extractor that computes an embedding for each speaker present in a speech mixture. To allow for supervised training, a teacher-student approach is employed: the teacher computes the target embeddings from each speaker's utterance before the utterances are added to form the mixture, and the student embedding extractor is then tasked to reproduce those embeddings from the speech mixture at its input. The system much more reliably verifies the presence or absence of a given speaker in a mixture than a conventional speaker embedding extractor, and even exhibits comparable performance to a multi-channel approach that exploits spatial information for embedding extraction. Further, it is shown that a speaker embedding computed from a mixture can be used to check for the presence of that speaker in another mixture.

arXiv commentProceedings of INTERSPEECH

References in corpus (8)

• ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification
• Deep Neural Networks for Multiple Speaker Detection and Localization
• Target-Speaker Voice Activity Detection: a Novel Approach for Multi-Speaker Diarization in a Dinner Party Scenario
• VoxSRC 2021: The Third VoxCeleb Speaker Recognition Challenge
• The DKU-DukeECE-Lenovo System for the Diarization Task of the 2021 VoxCeleb Speaker Recognition Challenge
• Tackling the Score Shift in Cross-Lingual Speaker Verification by Exploiting Language Information
• SJTU-AISPEECH System for VoxCeleb Speaker Recognition Challenge 2022
• Simple Attention Module based Speaker Verification with Iterative noisy label detection