Yanpei Shi

2 accepted papers

2021

Towards Low-Resource Stargan Voice Conversion Using Weight Adaptive Instance Normalization

ICASSP 2021accepted

Many-to-many voice conversion with non-parallel training data has seen significant progress in recent years. It is challenging because of lacking of ground truth parallel data. StarGAN-based models have gained attentions because of their efficiency and effective. However, most of the StarGAN-based w…

Cited by 0 Source ↗Scholar ↗

2020

H-Vectors: Utterance-Level Speaker Embedding Using a Hierarchical Attention Model

Yanpei Shi, Qiang Huang, Thomas Hain

ICASSP 2020accepted

In this paper, a hierarchical attention network is proposed to generate utterance-level embeddings (H-vectors) for speaker identification and verification. Since different parts of an utterance may have different contributions to speaker identities, the use of hierarchical structure aims to learn sp…

Cited by 0 Source ↗Scholar ↗