Factorization of Discriminatively Trained i-vector Extractor for Speaker Recognition
This work addresses computational bottlenecks for researchers and practitioners in speaker recognition, though it is incremental as it builds on existing i-vector methods.
The authors tackled the computational and memory inefficiency of the original generative i-vector model for speaker verification by factorizing it into a compact model with fewer parameters, achieving similar performance and further improving it through discriminative training to enhance results across various benchmarks.
In this work, we continue in our research on i-vector extractor for speaker verification (SV) and we optimize its architecture for fast and effective discriminative training. We were motivated by computational and memory requirements caused by the large number of parameters of the original generative i-vector model. Our aim is to preserve the power of the original generative model, and at the same time focus the model towards extraction of speaker-related information. We show that it is possible to represent a standard generative i-vector extractor by a model with significantly less parameters and obtain similar performance on SV tasks. We can further refine this compact model by discriminative training and obtain i-vectors that lead to better performance on various SV benchmarks representing different acoustic domains.