Zero-cost lip-sync video created with QwenTTS + Maestro

cocktailpeanut · x · 2026-08-19

The creator demonstrated a workflow using QwenTTS (running in Pinokio) to generate voice, followed by the Maestro app to animate a single reference image with the MP3 audio. The resulting lip-sync video was 100% free and involved no data centers in either step, highlighting an accessible local pipeline for AI-generated video.

Original post →

More from Multimodal

Multimodal channel →