Microsoft Unveils Open-Source Text-to-Speech Model, VibeVoice
2 week ago / Read about 0 minute
Author:小编   

Microsoft has officially launched VibeVoice, an open-source Text-to-Speech (TTS) model. Designed to produce expressive, extended, and multi-speaker conversational audio for applications like podcasts, VibeVoice aims to address the scalability issues, speaker voice inconsistency, and unnatural conversational transitions that plague traditional TTS systems.