Incremental Transfer Learning in Two-pass Information Bottleneck based Speaker Diarization System for Meetings
This incremental improvement reduces computational cost for speaker diarization in meeting recordings, benefiting real-time or large-scale applications.
The paper tackled the high real-time factor (RTF) in a two-pass information bottleneck speaker diarization system by using incremental transfer learning to update neural network parameters from previous conversations, reducing RTF by 33.07% on NIST RT-04Eval and 24.45% on AMI-1 datasets with minor performance degradation.
The two-pass information bottleneck (TPIB) based speaker diarization system operates independently on different conversational recordings. TPIB system does not consider previously learned speaker discriminative information while diarizing new conversations. Hence, the real time factor (RTF) of TPIB system is high owing to the training time required for the artificial neural network (ANN). This paper attempts to improve the RTF of the TPIB system using an incremental transfer learning approach where the parameters learned by the ANN from other conversations are updated using current conversation rather than learning parameters from scratch. This reduces the RTF significantly. The effectiveness of the proposed approach compared to the baseline IB and the TPIB systems is demonstrated on standard NIST and AMI conversational meeting datasets. With a minor degradation in performance, the proposed system shows a significant improvement of 33.07% and 24.45% in RTF with respect to TPIB system on the NIST RT-04Eval and AMI-1 datasets, respectively.