Vois 2.0 is live: local voice AI studio with 100+ voices, Voice Design, and founders pricing
Vois 2.0 is live.
This release moves Vois much closer to the product I wanted to build: a desktop voice AI studio for people making real audio, not just testing one prompt in a browser tab.
What changed:
100+ studio-quality voices
Local generation on your desktop
Scripts stay on your machine
No per-character meter on paid plans
Voice cloning from a consented sample on paid plans
Voice Design and Omni in Pro for broader voice creation and 600+ languages
Script workflow, retakes, mastering, and export in one app
I built this because cloud TTS pricing punishes iteration. Retakes, pacing fixes, pronunciation edits, and alternate reads should be part of the workflow, not something you ration.
Founders pricing is open this week for the first 100 founders:
Subscriber: USD 10/month with FOUNDERLOCAL
Pro: USD 14/month with FOUNDERPRO
Monthly only. Ends June 21. Founders keep the monthly rate while their subscription stays active.
If you try it, I would love honest feedback: where onboarding is confusing, which voices need work, and what would make Vois more useful for your workflow.
Replies
Voice Design sounds like a useful addition. Curious how much control users get over tone, age, emotion, and speaking style.
Vois
@freya24 great question.
Voice Design is descriptive rather than slider-only. In Pro, you can describe the kind of voice you want, including things like approximate age, tone, speaking style, energy, accent direction, and use case.
For example: a warm audiobook narrator, a sharper product-demo voice, a calm instructional voice, or a more dramatic character voice.
The goal is not just making a voice once, but making it usable inside the studio workflow: cast it into scripts, test lines, retake awkward reads, then master/export.
Curious what kind of voice you would want to create first?