To more intuitively demonstrate the performance of our method on the voice conversion task, we present some results of the any-to-one voice conversion task.
Any-to-one Voice Conversion aims to convert speech of any speaker to the speech with target speaker's voice.
The following audios are the original recordings of four different speakers.
The audio sample of the target speaker.
Wav2vec2.0 base
HuBERT base
WavLM base
ProgRE base
Wav2vec2.0 large
HuBERT large
WavLM large
ProgRE large