GR0: SELF-SUPERVISED GLOBAL REPRESENTATION LEARNING FOR ZERO-SHOT VOICE CONVERSION (ICASSP 2024)

Project Page

Index Page for GR0 Listening Samples

On this page, you will find a set of audio samples created by our method and baselines, which were benchmarked the listening test (MOS).

We present a single example per experiment here. For additional examples and the detailed explanations of the labels, please click on "more samples" to access a full list of the audio samples.

Voice Conversion

DAPS (unseen dataset, more samples)

VCTK (unseen dataset, more samples)

EN2PT (unseen dataset and language, more samples)

PT2EN (unseen dataset and language, more samples)