Skip to content
WorkflowListedExternal
Open source

Microsoft Vibe Voice Text To Speech Example Workflow — Updated

The example Workflow for the Vibe Voice Text to Speech AI models. Microsoft VibeVoice with ComfyUI is an AI tool that creates natural-sounding speech, supports voice cloning from audio samples, and can generate...

Sourcecivitai.com/models/1923801
TaskAudioEstimated VRAMNot yet characterizedBgrade

Runtime profile

Access
External workflow
Status
Listed
Source
Civitai
Updated
Oct 2, 2026

Source description

The example Workflow for the Vibe Voice Text to Speech AI models. Microsoft VibeVoice with ComfyUI is an AI tool that creates natural-sounding speech, supports voice cloning from audio samples, and can generate conversations with up to four distinct speakers, making it great for podcasts or long audio scripts. It offers options for controlling speech style and quality, working best with English and Chinese.

Link to ComfyUI Custom Node - https://github.com/wildminder/ComfyUI-VibeVoice

I have a free written manual installation guide that includes the required python packages and custom node links, available here - https://www.patreon.com/posts/137750868 .

If you want a full video walkthrough of the Gradio Demo version of the vibe voice project and installation process, including manual setup, recommended settings, and tips for both local and cloud deployment—check out my YouTube tutorial linked below:

What you need

Model files
0 GiB

Remaining hardware or download-size estimates are not available yet.

Requirements

Node packs

1

AI-generated commentary

AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.

Purpose

  • Use this ComfyUI workflow for text-to-speech: create natural-sounding speech from text, clone a voice from audio samples, and generate conversations with up to four distinct speakers for podcasts or long audio scripts.

Flow

  • The workflow includes LoadAudio, MarkdownNote, SaveAudio, and VibeVoiceTTS nodes.

Inputs and outputs

  • Inputs are text and, for voice cloning, audio samples; outputs are natural-sounding speech or conversations with up to four distinct speakers.

Setup

  • Setup materials include a manual guide covering required Python packages and custom-node links, plus a video walkthrough covering manual setup for local or cloud deployment.

Dependencies

  • The workflow requires the ComfyUI-VibeVoice node pack.

Configuration

  • You can control speech style and quality.

  • English and Chinese are the languages specifically described as working best.

Cautions

  • No license is provided for this workflow.

Does this need to be edited?

Sign in to send an edit request.

Sources

1 source
CivitaiPrimary
civitai.com/models/1923801

Verified Jul 19

Source excerpts

1 excerpt
NoteJul 19civitai.com/models/1923801

Source context: 699 downloads · Type Workflows · Base model Other