Speech utterances recorded under differing conditions exhibit varying de...
The emergence of large-margin softmax cross-entropy losses in training d...
This paper proposes the use of two task-aware warping factors in mask-ba...
We present a Bayesian formulation for deep speaker embedding, wherein th...
This paper presents a simple but effective method that uses multi-resolu...
This paper proposes a generalized framework for domain adaptation of
Pro...
The I4U consortium was established to facilitate a joint entry to NIST
s...
State-of-the-art speaker recognition systems comprise an x-vector (or
i-...
This paper presents an experimental study on deep speaker embedding with...