Resemble AI

Published September 3, 2026 · This page indexes what this site has verified about Resemble and links to where each finding is set out in full. It is not a review — the review is. No listening test has been run, so nothing here says how Resemble sounds.

On this page

What it costs

Resemble AI’s published rate card prices detection products only — Flex $0/mo, Team $350/mo ($280/mo billed annually, the vendor’s own wording "save $840/yr"), Business $1,000/mo ($800/mo billed annually, "save $2,400/yr") and Enterprise custom — with the comparison rows covering Resemble Detect, Intelligence, Meetings, Identity, Watermarker and platform/API access, and no speech-synthesis product in any row; the one usage rate on that page is Resemble Watermarker decode at "$0.00020 per call" (checked September 1, 2026, the review). A full-text search of that pricing page returned zero occurrences of "text-to-speech", "voice cloning", "speech synthesis", "voice generation" and "per character" (pricing comparison). The only synthesis figures the site could find sat in a May 8, 2026 changelog entry — "first clone free, then $2 each", with card capture deferred until after that first free clone — and a product FAQ, set against documentation gating the Voice Cloning API to the $1,000/month Business plan; as of September 2, 2026 the site records those as a record of what was once published rather than live rates.

Because no speech-synthesis rate is published, this provider carries no row in the normalized rate table of our pricing comparison; that page records the absence and why.

Where Resemble’s own material disagrees with itself

These are contradictions and documented gaps inside the vendor’s published material, not disagreements between us and the vendor. We report both readings rather than choosing one, and each links to the page where we set it out in full. We have published 14 in total; the 10 with the most bearing on a decision are below, and the rest sit in the pages linked underneath.

• Resemble’s homepage carried a badge reading "SOC 2 Type II — Independently audited security controls covering availability, confidentiality, and data integrity", while its trust centre stated that "Resemble AI is currently in our SOC 2 Type 2 observation period". Both were live on the same day, September 1, 2026, and an observation period is not a completed Type II audit. Where we published this.
• Resemble’s models index badges DramaBox as "MIT OPEN SOURCE", but the repository’s LICENSE file is not an MIT licence — it is the "LTX-2 Community License Agreement", licence date January 5, 2026, with Lightricks Ltd named as Licensor, reported by GitHub as SPDX NOASSERTION and by Hugging Face as "license: other", and its clause 2 requires entities with annual revenues of at least $10,000,000 to obtain a paid commercial use licence. A June 10, 2026 changelog entry records that Voice Design moved to the DramaBox model. Where we published this.
• The model the product page markets is the one the same company’s hosted API documentation retires: the developer docs state "Resemble Ultra is the current text-to-speech model for new and upgraded voices" and list Chatterbox (tts-v4), Chatterbox-Turbo (tts-v4-turbo) and Chatterbox Multilingual (tts-v4) as end of life, while the text-to-speech product page still leads with Chatterbox. Resemble Ultra itself has no product or model page — the only description found was an April 29, 2026 changelog line announcing it is live. Where we published this.
• Consent is marketed as guaranteed and written as optional: the ethics page says Resemble’s consent system "ensures voice cloning only occurs with the explicit consent of the individual", while the Terms of Service say "Resemble may require consent form the individual or third party whose voice is being cloned" — the typographical error is the vendor’s own. The site notes that "ensures" and "may require" are not the same commitment, and that the terms are the document that governs. Checked September 2, 2026. Where we published this.
• Two official answers on cloning price cannot be reconciled and the site publishes both rather than picking one: a changelog dated May 8, 2026 says "first clone free, then $2 each", while the developer documentation states "The Voice Cloning API requires a Business plan or higher" — a plan published at $1,000 a month with bullets that never mention voice. Where we published this.
• The documented entry price for programmatic cloning has to be assembled from two documents that do not reference each other: the docs state "The Voice Cloning API requires a Business plan or higher" and, for streaming, "This API is available to Business plans and above", while the $1,000/month Business plan’s own listed bullets — everything on Team, deployment configurability, calendar integration, security notifications, single sign-on, twenty seats — never mention voice, cloning or synthesis. Checked September 1, 2026. Where we published this.
• Official Resemble pages give several different language figures for the same offering: the text-to-speech product page cites 100 languages in three places with no enumeration published anywhere, the FAQ on that same page gives 23 with a named list, and the open-source repository README gives 23 with a different named list. Checked September 1, 2026. Where we published this.
• Counting the vendor’s own enumerated list against its advertised figures, the site records Resemble advertising both "25 languages" and "20+ languages" on the same page while the list enumerated on the model page came to 21 entries — three figures for one model. The 23-entry counts recorded above come from two other lists, the product-page FAQ and the repository README. Checked September 2, 2026. Where we published this.
• Resemble’s integrations and environments documentation advertises a "99.9% uptime SLA" for the cloud service, while its own status page on September 1, 2026 showed a 90-day figure of 99.425% for the Safety & Detection API, with Deepfake Detection at 99.805% and Identities at 99.044% (web portals showed 100% and 99.999%). Where we published this.
• Resemble’s API documentation contradicts itself twice on authentication — Bearer against a bare key on the same endpoint, and two different URLs for obtaining an API key — and publishes no error-code taxonomy, no documented 429 semantics, no Retry-After, no rate-limit response shape and no backoff guidance beyond a single WebSocket line. Checked September 2, 2026. Where we published this.

Dated and time-sensitive

Facts with a clock on them — model sunsets, policy revisions, market status. These are the entries most likely to be out of date first, which is why they carry their dates.

• 2026-08-13. Date of Resemble AI’s own machine-readable self-description, which states verbatim: "Resemble AI does not sell voice generation products and is not accepting new voice customers. Voice continues as open research only." The same document states it does not sell "text-to-speech, voice cloning, speech-to-speech conversion, or voice design as commercial products", that existing voice customers "continue to be supported indefinitely", and that characterizations of the company as a voice AI vendor, a TTS provider or a voice cloning platform "are out of date and describe a catalog the company no longer sells". The site read it on September 2, 2026. Source on this site.
• 2026-09-02. Correction published: Resemble AI is now unranked in the flagship comparison and removed from its ranked structured data, with a dated status notice added to nine pages — the review, the flagship ranking, the alternatives page, both comparisons naming it, and the cloning, free tier, premium tier and pricing comparisons. The affected pages were not deleted or quietly rewritten: their documented facts remain, dated, because existing customers are still governed by those terms. Source on this site.
• 2026-09-02. The MIT-licensed open-source models are still published and the licence is unchanged, but the vendor now describes them as "research artifacts rather than supported products" carrying "no SLA, paid tier, or enterprise support" — the frame the site’s pages now use. Source on this site.
• 2026-09-02. On the pricing benchmark, Resemble AI’s missing synthesis rate is now recorded as resolved rather than unverified: no rate is published because the vendor does not sell voice generation, and the changelog and FAQ figures are retained as a record of what was once published rather than as live rates. Source on this site.
• 2026-09-01. Model end-of-life on the hosted API: the developer documentation states that "All previous Resemble text-to-speech models have reached end of life. Voices that still use one of these versions can no longer generate audio and must be upgraded to Resemble Ultra", with Chatterbox (tts-v4), Chatterbox-Turbo (tts-v4-turbo) and Chatterbox Multilingual (tts-v4) listed in the end-of-life table. Source on this site.
• 2026-04-29. Changelog entry announcing that "Resemble Ultra (powered by xAI) is live" — the only description of the current hosted text-to-speech model the site could find, as it has no product or model page. Source on this site.
• 2026-05-08. Changelog entry titled "Voice cloning pricing: first clone free, then $2 each", stating that every user gets one clone before anything else and pays only from the second at $2 each, with card capture deferred until after that first free clone. Source on this site.
• 2026-06-10. Changelog entry recording that Voice Design moved to the DramaBox model — the model whose repository ships the LTX-2 Community License rather than the MIT licence its site badge claims. Source on this site.

Every page on this site about Resemble

• Full review. Resemble AI Review 2026: Pricing, Licensing and Watermarking, Docs-Verified — the complete documentation-based assessment, with sources and dates against every claim.
• Alternatives. Resemble AI alternatives — organized by the reason you would leave, not as a ranked list.
• Head to head. MiniMax vs Resemble AI, Resemble AI vs Speechify.
• In our buying guides. Best AI Voice Generators in 2026: 10 Providers Compared, Best AI Dubbing Software 2026: Only Four Vendors Actually Sell It, Best Free AI Voice Generators 2026: What “Free” Actually Buys You, Most Realistic AI Voices: What Every Vendor Charges for the Claim, Best Multilingual AI Voice Generators: Every Count, Recounted, Real-Time AI Voice: Why the Latency Numbers Cannot Be Compared, Best AI Voice Cloning Software 2026: Price, Access and Consent Compared.
• In our comparisons of published data. AI Voice Generator Pricing Comparison: Every Rate Normalized, Best Text-to-Speech APIs: Limits, Transports and Auth Compared.
• In our explainers. Can You Use AI Voices Commercially? What Ten Vendors Actually Say, How AI Voice Cloning Works: What Ten Vendors Require of You, How AI Voice Generators Work, According to the Vendors Themselves.
• By what you are making. AI Voice for Audiobooks: Will Chapter 20 Match Chapter 1?, AI Voice for Business: What Survives a Procurement Review, AI Voice for eLearning: Courses Change, and the Audio Must Follow, AI Voice for Podcasts: Only Three Providers Do Two Speakers, AI Voice for YouTube: What You Must Disclose Before You Publish.

What we could not verify

Honesty about gaps beats a page that looks complete. Standing limits on everything above, as of September 3, 2026:

• How Resemble sounds. Our listening benchmark is designed but not running. Nothing on this page or on any page it links to ranks Resemble on audio quality, naturalness or speed.
• Anything Resemble does not publish. Every page indexed here is built from the vendor’s own documentation. Where that documentation is silent, we record the silence rather than estimating around it.
• Whether a documented feature works as documented. We read documentation; we did not create an account or make an API call.
• Currency. Each linked page carries its own check date. This index inherits them and does not re-verify on load.

Change log

September 3, 2026 — First publication. Assembled from pages already published on this site; no new vendor claim was introduced by this page.

Published September 3, 2026 · This page shows no audio and reports no listening results, because our benchmark has not run. Corrections are recorded with their date on the corrections page. Independence note: this page contains no affiliate links, and no vendor paid for placement or influenced the order — see how we make money and our editorial policy.