Yingke Zhu

2 accepted papers

2020

Orthogonal Training for Text-Independent Speaker Verification

ICASSP 2020accepted

In this paper we propose orthogonal training schemes to improve the effectiveness of cosine similarity measurements in text-independent speaker verification (SV) tasks. Compared to the PLDA backend, cosine similarity is simple to compute, and it does not require extra data or time to build a separat…

Cited by 0 Source ↗Scholar ↗

2017

Speeding up softmax computations in DNN-based large vocabulary speech recognition by senone weight vector selection

Yingke Zhu, Brian Mak

ICASSP 2017accepted

Deep neural network has obtained significant accuracy improvement in many large vocabulary continuous speech recognition (LVCSR) tasks. Recently, it was shown that even better performance can be obtained by modeling a larger number of more discriminative senones. However, as the neural network becom…

Cited by 0 Source ↗Scholar ↗