Find tools › Open-source projects › Images, video & voice

VibeVoice

Open-source project · by microsoft

A tool that converts long audio recordings into written text with timestamps and creates realistic spoken voices.

For developersVery popular392 installs a month

What it costs

Free and open source; you may pay for the AI model it uses.

Use it to

  • Transcribe hour-long meeting recordings with speaker labels and timestamps
  • Convert continuous text into spoken voice audio in real time
  • Supply custom names and technical words to improve transcription accuracy

How to start

  1. Try the live demo online through Microsoft Foundry or Hugging Face spaces

  2. Select and download the speech recognition or real-time voice model weights from Hugging Face

  3. Run transcription scripts or stream audio using Python or BitNet edge binaries

Good to know. Microsoft removed the main multi-speaker long-form text-to-speech code from this repository due to responsible AI use concerns.

Before you install

  • Found automaticallyNot reviewed by a person yet. Check it before installing.
  • License: MITYou are allowed to use it under these terms.
  • Last updated 2026-10-08

More detail

VibeVoice is an open-source suite of voice models for long-form speech transcription and real-time voice synthesis.

Standard speech models struggle to transcribe long audio with multiple speakers or generate low-latency speech. Reach for VibeVoice to transcribe up to hour-long recordings with speaker names and timestamps or to generate real-time voice output.

Similar tools

Descriptions are written in plain English from each project's own documentation and refreshed by our agent; this one was last checked 2026-11-05. Spotted a mistake? · Open in Find tools · © 2026 Which AI Map