DNN based Speaker Recognition on Short Utterances
It addresses the challenge of reducing required speech length for speaker verification in real-world applications, showing incremental improvements over existing systems.
This paper tackles the problem of speaker verification with limited speech data by using a DNN-senone-based GPLDA system, achieving improvements of over 50% and 18% in EER values on specific NIST 2010 conditions compared to traditional methods.
This paper investigates the effects of limited speech data in the context of speaker verification using deep neural network (DNN) approach. Being able to reduce the length of required speech data is important to the development of speaker verification system in real world applications. The experimental studies have found that DNN-senone-based Gaussian probabilistic linear discriminant analysis (GPLDA) system respectively achieves above 50% and 18% improvements in EER values over GMM-UBM GPLDA system on NIST 2010 coreext-coreext and truncated 15sec-15sec evaluation conditions. Further when GPLDA model is trained on short-length utterances (30sec) rather than full-length utterances (2min), DNN-senone GPLDA system achieves above 7% improvement in EER values on truncated 15sec-15sec condition. This is because short length development i-vectors have speaker, session and phonetic variation and GPLDA is able to robustly model those variations. For several real world applications, longer utterances (2min) can be used for enrollment and shorter utterances (15sec) are required for verification, and in those conditions, DNN-senone GPLDA system achieves above 26% improvement in EER values over GMM-UBM GPLDA systems.