Abstract: Zero-shot text-to-speech (TTS) has recently achieved remarkable performance by leveraging a speech prompt instead of a speaker embedding, as it provides richer information. However, ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results