Choose the local voice studio that fits your work.
Start free with a 7-day trial, then subscribe or pay as you go. Cancel anytime.
Rated 5.0 on Product Hunt
$192/year, billed annually. Core local voice production without Omni or Voice Design.
- Unlimited audio generation
- 100+ studio-quality voices (and growing)
- Voice cloning from a short sample
- Script editor with multi-speaker support
- Multi-track timeline, mastering & EQ
- Audio export presets for Spotify, YouTube, and Apple Podcasts
- CLI access for automation
Cancel anytime · License key by email, instantly
$264/year, billed annually. $6/mo more than Subscriber for Omni, 600+ languages, and Voice Design.
- Everything in Subscriber, unlimited too
- Omni: the highest-quality local model
- One voice, 600+ languages
- Voice Design: new voices from a text description
- Cloned voices work across 600+ languages
- Pro quality and performance features
Cancel anytime · License key by email, instantly
Omni runs on your GPU. The free trial confirms it runs on your machine before you pay.
Try it free first
7-day trial. No card needed.
Pay as you go
1 credit = 1 export. No expiry.
At checkout, use the same email as your verified Vois account.
Omni requires a compatible graphics card. Check your computer before choosing Pro. All sales are final. See supported hardware.
Running a quick security check
The hidden cost of “affordable” cloud voices
Cloud TTS platforms sell credits by the character. Sounds cheap until you actually start producing.
- Previewing costs credits. Every time you press “Generate” to hear how a line sounds, credits are deducted. There is no free preview.
- Editing burns credits. Fix a typo, tweak punctuation, change one word. The entire line is re-charged at full price.
- Scale kills your budget. A 50,000-word audiobook runs ~300,000 characters. At $0.24/1K characters, that’s $72 per book, every time you iterate.
- No production tools. Generate audio, then switch to a separate editor to arrange, master, and export. Multiple apps, multiple subscriptions.
- Your script leaves your computer. Cloud synthesis works by uploading your text to their servers, so retention and access follow their policy, not yours.
- No connection, no audio. Hosted generation needs the internet, and their service, to be up when you want to work.
- Generate as much as you want. Preview, iterate, and experiment. No meter running, no anxiety about wasted credits.
- Edit freely. Change a word, re-generate a paragraph, try three different voices on the same line. Cost: $0 extra.
- Produce at any scale. Whether it’s one blog post or a 20-hour audiobook series, same flat price. No overages, no surprises.
- Complete production studio. Script editor, multi-track timeline, professional mastering, platform export presets, all built in.
- Your script stays on your computer. Speech generation and audio processing run locally. Your text and recordings are not uploaded to be turned into audio.
- Works offline. Once the app and your voices are installed, you can keep producing with no connection at all.
Real math: a 10-episode podcast season (~60,000 words)
Cloud TTS: ~360,000 characters × $0.24/1K = $86 per season (plus re-generation costs for edits).
Vois: $16/mo for unlimited seasons, unlimited edits, with a built-in production studio.
What you keep, on every plan
Revise without generation fees
Paid plans include unlimited local generation. Edit your text and generate another version without per-character charges.
Generate on your computer
TTS and mastering run on your machine. Scripts, reference audio, and generated audio are not uploaded for processing. Connected publishing remains user-controlled.
Keep your audio files
Cancel anytime. Every file you’ve generated stays on your computer. Your projects, your audio, your voices, yours forever.
Questions? Answers.
Plans, voice models, hardware requirements, and exports.