microsoft/VibeVoice
★ 54,881 stars
Covered in week 2026-W22 · the 2026-05 report
Open-Source Frontier Voice AI
What VibeVoice is
VibeVoice is a family of open-source frontier voice AI models that includes both Text-to-Speech (TTS) and Automatic Speech Recognition (ASR) models.
From the repository README · read it on GitHub →
Transcript
ElevenLabs charges fifty cents per minute for conversational voice AI. VibeVoice does it free. Microsoft's open-source model delivers real-time speech-to-speech with under 300 millisecond latency, streaming audio as you speak. Run it locally on your own GPU and build voice agents that listen, think, and reply out loud without paying per interaction. Two thousand stars already on GitHub in week one. No API keys, no usage caps, no monthly bill. Time to ditch ElevenLabs? Comment your take.