Price terms as published
Each line is the vendor’s own wording, split from the price capture checked on 2026-10-13. Nothing here is rounded, converted or re-worded — the full sentences stay in the review below. Lines that carry only a price continue the plan named on the line above, in the vendor’s own order.
| # | Plan / term | Checked |
|---|---|---|
| 1 | Free (1,000 one-time credits) | 2026-10-13 |
| 2 | Creator $16/mo billed yearly | 2026-10-13 |
| 3 | Studio $80/mo | 2026-10-13 |
| 4 | Scale $240/mo | 2026-10-13 |
One-liner: Onepin is an “agent layer for AI voice” — you bring a script, it plans the read, routes each line to the best of 30+ commercial TTS models, scores every generated line for accuracy (WER), naturalness, noise and pronunciation, and re-fixes or regenerates whatever fails before delivery.
Verified facts (checked 2026-10-13 against the official site)
| Attribute | Value | Source |
|---|---|---|
| Product shape | Script → normalize → route → generate → validate → fix pipeline | onepin.ai homepage + FAQ |
| Models | “30+ commercial models”, auto-routed per line; open-source models “not yet” | FAQ, 2026-10-13 |
| Languages | “10 languages today, with more coming” | FAQ |
| Voice cloning | “On the way” — not available yet | FAQ |
| Seats | Free on every plan (“you pay for the voice you generate, not headcount”) | FAQ |
| Credits | 100 credits = $1; generation + validation in one credit, “no second TTS invoice” | FAQ + pricing page |
| Free plan | 1,000 one-time credits (~15 min validated audio), 1 workspace, 1GB, 1 concurrent run, no download/export | pricing page |
| Paid tiers | Creator $16/mo, Studio $80/mo, Scale $240/mo (all billed yearly, 20% saving) | pricing page |
| Compliance badges | SOC 2 Type II claimed compliant; SAML SSO; 99.9% SLA with service credits | homepage + pricing page |
| GDPR | “We are working toward GDPR compliance” — not done | FAQ |
| Launch signal | Product Hunt .launch 2026-10-10, #2 that day | Product Hunt page (third-party) |
What it actually does
The pitch is that TTS engines make voice, but nobody makes it shippable. Onepin inserts a production line in front of the models you already pay for:
- Normalizer — rewrites numbers, dates, abbreviations and symbols into how they should be spoken (“$20” becomes “twenty dollars”). Onepin claims nearly half of TTS errors are fixed at this stage alone.
- Auto-routing — each line goes to whichever commercial model reads it best, informed by continuously-run cross-language benchmarks.
- Validation — every line is scored on word accuracy (WER), naturalness, noise and pronunciation before it ships; flagged lines are auto-fixed or regenerated.
- Fixes, not re-renders — mispronunciation corrections work at word level rather than regenerating the whole clip.
- Bring your own key — from Studio up, you can plug in your own ElevenLabs (and other provider) keys and run them through Onepin’s pipeline while keeping your provider rates.
The interesting commercial framing: it explicitly does not compete with ElevenLabs or OpenAI’s TTS. It sits above them and charges for the QA step.
Pricing detail (official pricing page, 2026-10-13)
| Free | Creator $16/mo | Studio $80/mo | Scale $240/mo | |
|---|---|---|---|---|
| Credits | 1,000 one-time | 3,000/mo | 15,000/mo | 50,000/mo |
| Overage | none | $0.010/credit | $0.008/credit | $0.007/credit |
| Concurrent runs | 1 | 3 | 5 | 10 |
| Download/export | not included | included | included | included |
| BYOK | — | — | included | included |
| Storage | 1GB | 100GB | 500GB | 1TB |
| Workspaces | 1 | — | 3 | 5 |
Enterprise (custom) adds SAML SSO, DPAs & BAAs, negotiated IP indemnity and pooled credits. Two billing facts worth pre-empting: unused credits do not roll over on self-serve plans (each cycle starts fresh), and the Free tier’s “in-app playback is forever free” is doing real work — the paywall is on taking the file out.
The fine print
- 10 languages, not 30. Translation ambitions aside, current supported languages are listed as 10 by the FAQ itself; they’ll prioritize more on request.
- GDPR is in progress by their own words — a blocker for some EU procurement checklists, even with SOC 2 Type II already claimed.
- Open-source model support is “coming soon” — if your plan was “route to Kokoro/XTTS for free”, that’s not today.
- Third-party writeups mention US/UK-English-only and beta status for word-level pronunciation fixes; the official FAQ we captured doesn’t state this either way, so treat that specific limitation as unconfirmed until they document it.
Who it’s for
Teams shipping volume voice content — audiobooks, course narration, VTuber/YouTube pipelines, ads — where one mispronounced brand name in a hundred clips is a re-render-and-reupload nightmare. The free 15 minutes lets you test the validator on your own worst-case script; export requires Creator. If you just need a voice, any single TTS is simpler. If you need the same script to come out clean across 500 lines and three providers, this is the category it’s inventing.
All figures above were captured from onepin.ai’s own pricing page and FAQ on 2026-10-13; raw captures archived alongside this entry. We have not run the product ourselves — this entry is marked official-sources review, and carries no score for exactly that reason.