VibeVoice on a Mac

Microsoft's long-form, multi-speaker TTS · Speech AI

VibeVoice is Microsoft's open-source frontier text-to-speech model for long-form, multi-speaker audio; expressive, podcast-style conversations with up to four distinct speakers and natural turn-taking. It pairs an LLM with a diffusion head and a low-frame-rate acoustic tokenizer to stay coherent over long durations. Run it locally from the GitHub repo / Hugging Face weights (a 1.5B model and larger variants; a GPU helps): install the package, then synthesize from a script of speaker turns. Good for narration, dialogue and synthetic voice-over. Weights are MIT-licensed and for research use.

Run VibeVoice with FrontierStack

Install or connect VibeVoice; then check its status, ports and certificate from FrontierStack. The same screen links to firewall checks, Malware Audit and backups where they apply.

Related in Speech AI

• Handy
• Wispr Flow
• Voicebox
• Scriberr
• Whisper
• Whisper.cpp
• Faster Whisper
• Coqui TTS