Voice-Cloned AI Singing Still Out of Reach: Testing LTX, MiniMax, AceStep
why_not_zoidberg_82 · reddit · 2026-09-11
A Reddit user tested whether any music-generation tool can clone a voice reference and sing properly. LTX 2.3 makes simple songs but lacks voice cloning; its In-Context LoRA clones voices but produces acapella-like results. MiniMax H3 ref2va only speaks lyrics, and AceStep 1.5 accepts music references without voice cloning. Conclusion: no current tool combines voice cloning with full song generation.
More from Multimodal
- Midjourney srefs behave differently by aspect ratio — 1:3 nails subtle painterly styles — Kyrannio · 2026-09-11
- "I gave my computer synesthesia" — a delightful real-time demo — floguo · 2026-09-11
- AI movie studio builds 3D camera previs tool to save generation credits — purecharisma2020 · 2026-09-11
- Tutorial released for the user-friendly MiniMax H3 workflow — roychodraws · 2026-09-11
- Google brings 41 accepted papers to ECCV 2026, from video spatial understanding to TIPSv2 — ymatias · 2026-09-11
- Creator directs AI-generated music video for her own cover song — sofiafarhan · 2026-09-11