# Inworld Instant Voice Cloning > Inworld's Instant Voice Cloning API makes a reusable text-to-speech voice from 3 to 30 seconds of audio in one synchronous call, for Inworld's TTS-2 models. The same operation is in the Inworld CLI and its bundled MCP server. - Canonical: https://www.anchorterminal.com/tools/inworld-voice-cloning - Pricing: https://www.anchorterminal.com/tools/inworld-voice-cloning#pricing - Reviews: https://www.anchorterminal.com/tools/inworld-voice-cloning#reviews - Markdown: https://www.anchorterminal.com/tools/inworld-voice-cloning.md (~9,050 tokens) - Slim: https://www.anchorterminal.com/tools/inworld-voice-cloning.min.md (~2,080 tokens, same facts, less prose, for token-sensitive contexts) - JSON: https://www.anchorterminal.com/tools/inworld-voice-cloning.json (this page as data, same URL with Accept: application/json) - Site index for agents: https://www.anchorterminal.com/llms.txt (the key docs and every listing in one file, at most 16 MB, about 4 million tokens: https://www.anchorterminal.com/llms-full.txt) - API: https://www.anchorterminal.com/api/v1/index.json - MCP: https://www.anchorterminal.com/mcp (the whole catalogue as tools, no key) - DNS: dig +short TXT inworld-voice-cloning.grade.anchorterminal.com (the grade as a DNS TXT record, for a sandbox with no HTTP where names still resolve; also top., best.. and .vs.. under the zone, and the same under grade.letme.dev) - Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data) - Updated: 2026-11-07 ## Overview **Grade C · 61.9/100 · rank #585 of 1486 · #4 in Voice cloning & custom voices · not agent-ready · confidence medium** More from Inworld AI, listed separately because each is its own product: [Inworld STT](https://www.anchorterminal.com/tools/inworld-stt.md) (Speech-to-text), [Inworld Realtime TTS-2](https://www.anchorterminal.com/tools/inworld-tts.md) (Text-to-speech), [Inworld Realtime API](https://www.anchorterminal.com/tools/inworld-realtime.md) (Conversational voice agents). ## Assessment One JSON call returns a usable voice ID from 3 to 30 seconds of audio, with no charge per clone and 100 custom voices on the free plan, behind keys that are read-only until write access is switched on. The API has no consent field, and Zero Data Retention excludes cloning samples. ## Facts | Field | Value | | --- | --- | | Vendor | Inworld AI (https://inworld.ai) | | Kind | Model API | | Category | Voice cloning & custom voices (https://www.anchorterminal.com/categories/voice-cloning) | | Transport | HTTP, stdio, Streamable HTTP | | Endpoint | `https://api.inworld.ai` | | Auth | API key · Self-serve workspace API key from the Portal or `inworld workspace add-key`, sent as `Authorization: Basic` with the Base64 credential as copied. Keys are read-only for the Voice API by default, and cloning, updating and deleting voices need the write toggle, which can be changed on an existing key. Deleting a key revokes it. One-time tokens minted from a key serve browser and mobile clients. | | Pricing | Freemium ($25 / mo) · Creating an instant clone has no charge in the docs or on the pricing page. Speech from a clone costs $25 per million characters on TTS-2 or $15 on TTS-2 Flash on the free On-Demand plan, falling on the $25, $100, $300 and $1,500 monthly plans. On-Demand allows 100 custom voices and 2 cloning requests a minute (checked 2026-11-06). | | x402 | No · No x402, MPP or L402 in the docs, llms.txt or the pricing page (checked 2026-11-06). Credits are bought at a checkout page, $10 minimum. | | Licence | Proprietary service under Inworld's Terms of Service. The `@inworld/cli` and `@inworld/tts` npm packages and the examples repository are MIT | | Tools exposed | 40 | | Packages | npm: `@inworld/cli`; npm: `@inworld/tts`; pypi: `inworld-tts` | | Source | https://github.com/inworld-ai/inworld-api-examples | | Docs | https://docs.inworld.ai/tts/instant-voice-cloning | | llms.txt | https://docs.inworld.ai/llms.txt | | Last release | 2026-10-29 | | npm downloads / week | 3,632 | | Surface listed | Inworld's Voice API cloning endpoints at https://api.inworld.ai, read from docs.inworld.ai, the pricing and terms pages on inworld.ai, the status page's GitHub issues and the published `@inworld/cli` 1.5.0 package on 6 November 2026. No account was created and no API call was made | | Sample length | 3 seconds minimum, 30 seconds used. The reference recommends 10 to 30 seconds. WAV or MP3 through the API, also webm in the Portal, 12 MB maximum | | Instant or professional | Instant is synchronous and generally available. Professional Voice Cloning is Beta, needs 10 minutes of audio and trains asynchronously | | Voice design | Yes, `POST /voices/v1/voices:design` from a text description, then `:publish`. Same rate limits as cloning | | Consent and verification | The Portal asks the user to confirm rights to clone the voice. The API request has no consent or verification field. No watermarking found in the docs | | Voice ownership | The docs say the creator stays owner of a voice shared to the Voice Library and grants Inworld a licence to host and distribute it. The Terms of Service (11 June 2025) have no clause specific to cloned voices or consent | | Endpoints | `POST /voices/v1/voices:clone`, `GET /voices/v1/voices`, `GET`, `PATCH` and `DELETE /voices/v1/voices/{voiceId}`, `GET /voices/v1/voices:customVoiceCapacity`, and copy tokens to move a voice between workspaces, on `https://api.inworld.ai` | | Models | Clones work on `inworld-tts-2` and `inworld-tts-2-flash` | | Rate limits | Voice cloning API requests a minute. On-Demand 2, Creator 3, Builder 5, Developer 10, Growth 20, Enterprise custom, per account across all keys | | Credentials | Workspace API key sent as HTTP Basic. Voice write access is an opt-in toggle per key. Deleting a key revokes it and every token minted from it | | Free tier | On-Demand plan is free with up to 70 minutes of TTS and 100 custom voices, and the docs say instant cloning is open to all users. Professional voice clones need a payment method on file | | Data retention | Cloning samples are kept to build and support the voice. Zero Data Retention does not cover them. No retention period published in the docs | | Regions | US by default. Isolated EU and India deployments for Enterprise, with voices and keys separate per region | | SDKs and tools | `@inworld/tts` 1.1.1 (npm, MIT, 6 May 2026), `inworld-tts` (PyPI), `@inworld/cli` 1.5.0 with the `inworld-mcp` server (npm, MIT, Node.js 20 or later) | | Custom voices | Cloned and designed voices share one limit per account. On-Demand 100, Creator 500, Builder 3,000, Developer 10,000, Growth 30,000, Enterprise custom, per the pricing page | | Capabilities | voice.clone, voice.design, speech.tts | | Tags | hosted, official, freemium, free-tier, no-card, closed-source, python, typescript, llms-txt, mcp, cli, status-page, soc2 | | JSON | https://www.anchorterminal.com/api/v1/tools/inworld-voice-cloning.json | ## Score breakdown (methodology v0.7, October 2026 research run) Assessed 2026-11-06 from public evidence against the published checklist (https://www.anchorterminal.com/benchmark/#checklist). Confidence: medium. Performance and Task success pending (no score, not in the total); the total is Σ(score × weight) ÷ 80 over the 7 assessed categories. "This run" is each category's share of the 100 points. | Category | Weight | This run | Score (0–100) | Points | | --- | --- | --- | --- | --- | | Reliability | 16% | 20 | 75 | 15.0 | | Performance | 10% | pending | pending | n/a | | Schema & documentation | 13% | 16.2 | 64 | 10.4 | | Agent ergonomics | 13% | 16.2 | 64 | 10.4 | | Security & auth | 14% | 17.5 | 55 | 9.6 | | Payments & pricing | 10% | 12.5 | 40 | 5.0 | | Task success | 10% | pending | pending | n/a | | Maintenance & community | 7% | 8.8 | 68 | 6.0 | | Transparency & trust (editorial 51, provenance 75) | 7% | 8.8 | 63 | 5.5 | | Negative events | up to −15 | up to −15 | none recorded | 0 | | **Total** | | | | **61.9 → C** | ### Why each score - Reliability 75 (checked 2026-11-06): Graded on the REST API, the surface an agent would call. Status page at status.inworld.ai, drawn from GitHub issues, with five components and none for cloning (20). Read through the GitHub API on 6 November, the 90 days from 8 August hold two incidents, 29 minutes of elevated TTS latency on 25 August and 26 minutes of elevated TTS 2.1 errors from a network fault at one GPU site on 28 October, both minor (20). Cloning limits are published per plan, 2 to 20 requests a minute (15). A 429 is documented with exponential backoff and jitter code, and since the 9 October check the rate-limits page says limits over a minute can carry `Retry-After`. No idempotency key or safe-retry guidance for clone creation (10 of 15, was 9). The pricing page, now read, names an SLA on Enterprise only with no terms or figure, as on inworld-tts (0). Instant cloning carries no Preview or Beta mark (10). 20+20+15+10+0+10 = 75. - Performance: Pending. Latency is measured per call by our probes, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until the first probe window closes. - Schema & documentation 64 (checked 2026-11-06): No OpenAPI file is linked from the docs or `llms.txt`, re-read on 6 November. The MCP server has typed inputs on all 40 tools, which earns part of the line (10 of 25). `llms.txt`, `llms-full.txt` and Markdown twins (10). The guides say when to use instant cloning, Professional Voice Cloning or Voice Design, and the reference warns which fields are ignored (15 of 20). Required fields and base64 audio are typed, and a new `languageCode` string is validated against the catalogue with `INVALID_ARGUMENT`, while size and length limits sit in prose and the MCP tool still states different ones (10 of 15). Examples in curl, Python, JavaScript and the CLI, with one error response shown and no error list for the Voice API (9 of 15). A `/v1/` path and dated TTS release notes, newest 11 September, with no changelog for the Voice API itself (10 of 15). 10+10+15+10+9+10 = 64. - Agent ergonomics 64 (checked 2026-11-06): The clone response echoes up to 15 seconds of processed audio as base64 with no way to leave it out. The MCP tool returns a compact result, but the server loads 40 tools with no toolsets, registering only those the credentials can call (13 of 25). Voice lists take an AIP-160 `filter`, `pageSize`, `pageToken` and `orderBy`, including `source = "IVC"`, and a capacity endpoint reports used and maximum custom voices (20). Errors are a numeric code, a message and details, and the one example reads clearly, but no list is documented for the Voice API (12 of 20). No idempotency key, no MCP annotations in the 1.5.0 package, and the SDKs retry 5xx and network errors twice by default (4 of 20). Two required fields, language auto-detection, and official Python and Node.js SDKs (15). 13+20+12+4+15 = 64. - Security & auth 55 (checked 2026-11-06): Graded for voice cloning as the other listings in this category were, with consent and misuse controls in place of the read-only line and training and retention of voice data in place of the prompt-injection line. Revocable workspace keys that are read-only for the Voice API until a write toggle is set, Realtime-only keys and single-use tokens, sent in a header (25 of 30). The API has no consent or verification field and no watermark was found. Only the Portal asks the user to confirm rights, and the Terms (11 June 2025), now read, have no clause on cloned voices. The write toggle keeps a default key from cloning (6 of 20). The Terms say Inworld won't train generally available models on non-public Materials except feedback and trust and safety content, while the Privacy Notice lists training among its uses. Zero Data Retention excludes cloning samples and no retention period for them is published (7 of 15, was 4). Usage is attributed per key in the dashboard and an experimental usage records endpoint exists from the Builder plan. No audit log found (7 of 15). GDPR and SOC 2 Type II are stated on every plan on the pricing page, and the security page links a vulnerability report form. No security.txt (404 per the 30 October inworld-tts check) and no bug bounty found (10 of 20, was 8). 25+6+7+7+10 = 55. - Payments & pricing 40 (checked 2026-11-06): No x402, MPP or L402 (0). The pricing page, unread on 9 October and read on 6 November, publishes per-character rates for speech from a clone, $25 per million characters on TTS-2 and $15 on TTS-2 Flash on the free plan, and no charge for creating a clone (20, was 10). The On-Demand plan is free with 100 custom voices and instant cloning, and no card is asked for (20). `inworld login` opens a browser and signup is a human flow. The CLI mints keys only after that (0). 0+20+20+0 = 40. - Task success: Pending. Task success needs the category task suites run through each tool, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until then. A data provider's data-quality score is published on its listing now and becomes half of this category when it's scored. - Maintenance & community 68 (checked 2026-11-06): Read as a model, on the lines inworld-tts and cartesia-voice-cloning use (recency 30, dated releases 10, deprecation notice 10, support 15, SDKs 15, health 10). The 9 October draft used the tool lines and gave 75. `@inworld/cli` 1.5.0, whose MCP server carries `clone_voice`, on 29 October 2026, after 1.4.0 (13 October), 1.4.1 (23 October), 1.4.2 and 1.4.3 (25 October) (30). Dated TTS release notes on 9 August and 11 September 2026, and none for the 28 October model update or for the Voice API (10). `inworld-tts-1` was discontinued on 15 June 2026 with requests routed to newer models, and clones made before 15 August keep a 15-second prompt, with no notice period stated (5 of 10). A public changelog, a Discord server and support@inworld.ai, with no evidence read of how they answer (8 of 15). `@inworld/tts` 1.1.1 dates from 6 May 2026 (10 of 15). The CLI is current, but the repositories named in its npm metadata aren't public (5 of 10). 30+10+5+8+10+5 = 68. - Transparency & trust 63 (checked 2026-11-06): Closed service with clear terms naming Theai, Inc. dba Inworld AI, read on 6 November (unread on 9 October because of the Content-Signal line), and MIT client packages on npm (15 of 30, was 8). The Zero Data Retention page is specific that cloning samples are retained to build the voice, but gives no period, and the Terms' no-training clause and the Privacy Notice's training use sit side by side without a reconciling sentence, as on inworld-tts (14 of 30). Dated notices in the docs, `inworld-tts-1` discontinued on 15 June 2026 and the 30-second clone limit from 15 August, with release stages defined and no written notice period (12 of 20). US by default with isolated EU and India regions on Enterprise. No sub-processor list was found in the docs (10 of 20). 15+14+12+10 = 51. Fix list, everything this grade says the listing lacks, the biggest gain first (18 items). What it covers is at https://www.anchorterminal.com/fixes/inworld-voice-cloning.md (JSON https://www.anchorterminal.com/fixes/inworld-voice-cloning.json). The whole list for a coding agent comes from `POST https://www.anchorterminal.com/api/v1/fix-list` with `{"listing": "inworld-voice-cloning", "plan": "read"}`, free while the beta runs ### What we couldn't check - Changed since the 9 October draft (held then only for Content-Signal ai-input=no). The pricing page and terms were read, no charge per clone and 100 free custom voices were confirmed, the 429 guidance now mentions Retry-After, `@inworld/cli` moved from 1.3.1 to 1.5.0 with the same 40 tools, one more minor incident (28 October), and the slug is inworld-voice-cloning in place of inworld-instant-voice-cloning, matching cartesia-voice-cloning and the other listings in the category. - unchecked: whether cloning samples are used to train models. The Terms rule out training on non-public Materials with two exceptions, and the docs don't say how that applies to samples - unchecked: the trust centre for the SOC 2 report and any sub-processor list - The repositories named in npm metadata (inworld-ai/inworld-cli, inworld-ai/inworld-tts-node) are not public, so the MCP tool definitions were read from the published 1.5.0 package and not run - Whether the API size limit for a sample is 12 MB as the Portal guide says or 4 MB as the MCP tool says - inworld.ai/terms is the version effective 11 June 2025, while ultravox.ai carries Inworld AI Terms effective 1 October 2026 after Inworld took over Ultravox (per the inworld-tts check). Which version governs wasn't established. ### Sources - instant voice cloning guide: (seen 2026-11-06) - clone endpoint reference: (seen 2026-11-06) - API keys and write permissions: (seen 2026-11-06) - rate limits, 429 and Retry-After: (seen 2026-11-06) - plans, cloning limits by plan and character counting: (seen 2026-11-06) - pricing, custom voices per plan, SLA row: (seen 2026-11-06) - Terms of Service, 11 June 2025: (seen 2026-11-06) - zero data retention scope: (seen 2026-11-06) - TTS release notes: (seen 2026-11-06) - CLI, MCP server and telemetry: (seen 2026-11-06) - docs index for agents: (seen 2026-11-06) - status incidents as GitHub issues: (seen 2026-11-06) - TTS incident, 28 October 2026: (seen 2026-11-06) - CLI versions and the MCP tool definitions in the 1.5.0 package: (seen 2026-11-06) - robots.txt with the Content-Signal line: (seen 2026-11-06) ## Score history Every move of a score since the listing was graded, newest first, with its cause. A score can't move without an entry here. JSON: https://www.anchorterminal.com/api/v1/tools/inworld-voice-cloning/history.json - 2026-11-06: graded 61.9 C (methodology 0.7) ## Who's behind it (provenance 75/100, checked 2026-11-06) | Check | Finding | Points | | --- | --- | --- | | Legal entity named | Theai, Inc. dba Inworld AI | 20/20 | | Domain age | inworld.ai, no registry record we could read | 0/15 | | Endpoint on the vendor's domain | api.inworld.ai | 15/15 | | Terms of service | read, states 6 of the 7 things a reader expects | 9.1/10 | | Privacy policy | read, states 7 of the 8 things a reader expects | 9.3/10 | | Status page | status.inworld.ai | 10/10 | | Changelog | published | 10/10 | | security.txt | not found | 0/10 | | People named | 3 named with their role, none with a public profile | 7/10 | | Support route | a support address, help pages, a community | 5/5 | The Terms of Service (effective 11 June 2025, re-read 6 November 2026) are an agreement with Theai, Inc. dba Inworld AI under California law. They carry no clause on robots or automated access, and none specific to voice cloning. inworld.ai's robots.txt carries `Content-Signal: ai-train=no, search=yes, ai-input=no`, recorded as a fact about the site. Since 30 October 2026 this is not a reason to hold, so this listing, held on 9 October for it, is graded. status.inworld.ai draws incidents from GitHub issues in inworld-ai/inworld-issue-status, read through the GitHub API on 6 November. inworld.ai/.well-known/security.txt returned 404 on 30 October per the inworld-tts check. The security page links a vulnerability report form. ### Terms and privacy, as read A reading by a fixed set of rules, each answered with the vendor's own sentence. Not legal advice. **Terms of service** (https://inworld.ai/terms), read 2026-10-30, dated 2025-06-11, states 6 of the 7 things a reader expects. - Gives the date it was last updated. Last updated 2025-06-11. - Names the governing law or courts. Disputes go to the courts of Santa Clara, California. - States a limit on its liability. Capped at the greater of $100 and the fees paid in the 12 months before the claim. - Says how changes to the terms are announced. Gives 30 days of notice before a change. - Not found in the text. Refers to a service level or uptime commitment. **Privacy policy** (https://inworld.ai/privacy), read 2026-10-30, dated 2026-01-18, states 7 of the 8 things a reader expects. - To know. Says it sells personal data or shares it for advertising. "Some of the activities described in this section may constitute “targeted advertising,” “sharing,” or “selling” under certain laws." - Gives the date it was last updated. Last updated 2026-01-18. - Says how long data is kept. For as long as needed, with no period named. - Says whether personal data is sold or shared for advertising. Says it does not sell personal data. - Gives a privacy contact. privacy@inworld.ai. - Not found in the text. Says where data is transferred or stored. ## Humans behind the project (researched 2026-11-06 from public pages, not part of the grade) ### Key people - Kylan Gibbs, Co-founder and CEO - Ilya Gelfenbeyn, Co-founder - Michael Ermolenko, Co-founder - Note: Copied from the inworld-tts humans file (checked 30 October 2026), the same company, on 6 November 2026 without new research. Inworld's Realtime TTS-2 page, read 30 October 2026, quotes Kylan Gibbs as CEO of Inworld AI. Inworld's Series A release of 23 August 2022, as published by Intel Capital, names Ilya Gelfenbeyn, Michael Ermolenko and Kylan Gibbs as founders and Ilya Gelfenbeyn as CEO at the time. No about or team page was found in the inworld.ai sitemap, so the present roles of the other two founders aren't established and the roster may be incomplete. No personal accounts are linked from Inworld's pages, so none are listed. The 28 October 2026 Reasoning post is signed by the Inworld TTS team. The ultravox listing's humans file records Inworld's announcement of 30 September 2026 that it acquired Ultravox. Investors come from the Series A release as published by Intel Capital and from The SaaS News report of the August 2023 round led by Lightspeed. Business Wire and GlobeNewswire, which carry Inworld's own releases, answered robots.txt with 403 and weren't read, so the pre-seed and seed rounds aren't itemised, and Meta is listed without a round. TechCrunch's report of 2 August 2023, read 6 November 2026, says Lightspeed invested more than $50 million in the latest tranche, "bringing Inworld’s total raised to more than $100 million at a $500 million post-money valuation", and mentions investments from Disney as part of the 2022 Disney Accelerator; that gives the valuation, the total raised and Disney below. The same report's grant from Epic isn't equity and isn't listed. No later round was found. On 6 November 2026 the ownership and investors were aligned with inworld-realtime, sourced to GamesBeat, since techcrunch.com's robots.txt disallows ClaudeBot. - Inworld AI (Theai, Inc.): LinkedIn https://www.linkedin.com/company/inworld-ai X https://x.com/inworld GitHub https://github.com/inworld-ai ### Ownership and investors - Raised: over $100M in total, as of August 2023 (source https://gamesbeat.com/inworld-ai-raises-new-round-at-500m-valuation-for-ai-game-characters/) - Valuation: over $500M as announced (the source doesn't say pre- or post-money), August 2023 round led by Lightspeed, August 2023 (source https://gamesbeat.com/inworld-ai-raises-new-round-at-500m-valuation-for-ai-game-characters/) - Backed by: Section 32, Intel, Lightspeed Venture Partners, Founders Fund, Accelerator Investments LLC, First Spark Ventures, Kleiner Perkins, BITKRAFT Ventures, CRV, Microsoft, Micron, LG, SK Telecom, The Venture Reality Fund, HTC, Meta, Stanford University, Samsung, The Walt Disney Company, NTT DOCOMO ### Getting support - Discord: https://discord.gg/inworld - Support email: support@inworld.ai - Help: https://docs.inworld.ai/portal/support - Help: https://docs.inworld.ai/ - Sources: https://inworld.ai/blog/realtime-tts-2, https://inworld.ai/blog/realtime-tts-2-reasoning-speed-consistency, https://inworld.ai/pricing, https://inworld.ai/terms, https://inworld.ai/security, https://inworld.ai/sitemap.xml, https://docs.inworld.ai/portal/support.md, https://www.intelcapital.com/inworld-ai-closes-50m-series-a-to-populate-video-games-and-the-metaverse-with-interactive-ai-driven-characters/ ## Live (updated 2026-11-07 18:20 UTC) - Right now: up, HTTP 200, 178 ms, checked 2026-11-07 18:20 UTC (get on `https://api.inworld.ai`) - Uptime 24h 100.0% (200 probes) · 30 days 100.0% (200 probes) · p50 136 ms · p95 266 ms - npm `@inworld/cli` 1.5.0 - npm `@inworld/tts` 1.1.1 - pypi `inworld-tts` 1.2.1, released 2026-05-06 - security.txt: none - Always current: https://www.anchorterminal.com/api/v1/live/inworld-voice-cloning.json ## Probe metrics Not measured yet. Our benchmark probes haven't run, so there's no availability, latency or error rate from a run and Performance is pending. Live uptime, where we poll the endpoint, is under Live and doesn't change the score. ## Prices | Item | Price | Unit | Note | | --- | --- | --- | --- | | Speech from a clone, TTS-2, On-Demand | $25 | per 1M characters | $20, $17.50, $15 and $12.50 on the monthly plans | | Speech from a clone, TTS-2 Flash, On-Demand | $15 | per 1M characters | $10 down to $7 on the monthly plans | | Creator plan | $25 | per month (plan) | credits equal to the price, 500 custom voices, 3 clone requests a minute | | Builder plan | $100 | per month (plan) | 3,000 custom voices, 5 clone requests a minute | | Developer plan | $300 | per month (plan) | 10,000 custom voices, 10 clone requests a minute | | Growth plan | $1500 | per month (plan) | 30,000 custom voices, 20 clone requests a minute | Across all listings: https://www.anchorterminal.com/prices/index.md ## What a task costs (from the prices above) | Task | Cost | At | | --- | --- | --- | | Narrate a 9,000-character script (about 1,500 words) | $0.13 to $0.22 | Speech from a clone, TTS-2 Flash, On-Demand to Speech from a clone, TTS-2, On-Demand | The quantity times this listing's own price, from its cheapest item in that unit to its dearest. Plans, minimums and free allowances aren't counted. ## Strengths - Synchronous clone from 3 to 30 seconds of WAV or MP3 audio, with language auto-detection and optional background-noise removal - No charge per clone. The free On-Demand plan includes instant cloning and 100 custom voices, and speech is $25 per million characters on TTS-2 - API keys are read-only by default. Cloning, updating and deleting voices need a write toggle that can be changed on an existing key - `GET /voices/v1/voices` filters by `source = "IVC"` with paging, and `voices:customVoiceCapacity` reports how many custom voices remain - `llms.txt`, `llms-full.txt` and a Markdown twin of every docs page, plus `clone_voice` in the MCP server in `@inworld/cli` 1.5.0 (29 October 2026) ## Weaknesses - No consent, verification or watermark field in the clone API. Only the Portal flow asks the user to confirm rights to the voice - Zero Data Retention does not cover cloning audio samples, and no retention period for them was found - No idempotency key on `POST /voices/v1/voices:clone`, and no error list for the Voice API - No OpenAPI file is linked from the docs, and the repositories named in the npm metadata of the CLI and Node SDK are not public - The MCP `clone_voice` description says about 5 seconds and 4 MB a sample, while the docs say 3 seconds and 12 MB ## Before you call it (notes for agents) 1. Create the key with write access (`inworld workspace add-key --write`). A default key gets code 7, insufficient access level, on `voices:clone` 2. Send the complete Base64 credential as `Authorization: Basic $INWORLD_API_KEY`. Don't encode it again 3. Send one 10 to 30 second WAV or MP3 clip as base64 in `voiceSamples[].audioData`. Audio past 30 seconds is cut 4. Don't blind-retry a clone. There's no idempotency key, so list voices with `filter=source="IVC"` first, and back off with jitter on 429 5. Set `gender`, `ageGroup` and `categories` with UpdateVoice afterwards. The clone call ignores them without an error 6. The response echoes up to 15 seconds of processed audio as base64 in `audioSamplesValidated`. Drop it before passing the result to a model ## Connect Install: ```bash npm install -g @inworld/cli ``` First request: ```bash curl --location 'https://api.inworld.ai/voices/v1/voices:clone' \ --header "Authorization: Basic $INWORLD_API_KEY" \ --header 'Content-Type: application/json' \ --data '{"displayName": "my_voice_clone_demo", "languageCode": "en-US", "voiceSamples": [{"audioData": ""}]}' ``` Claude Code: ```bash claude mcp add inworld -- inworld mcp ``` MCP client configuration: ```json { "mcpServers": { "Inworld": { "args": [ "mcp" ], "command": "inworld" } } } ``` ## Similar tools Ranked by shared capabilities, then score. Same-category tools with no shared capability key are listed last. | Tool | Grade | Score | Rank | Shared capabilities | x402 | Markdown | | --- | --- | --- | --- | --- | --- | --- | | ElevenLabs Voice Cloning and Voice Design API | BB | 76.8 | 40 | voice.clone, voice.design, speech.tts | no | https://www.anchorterminal.com/tools/elevenlabs-voice-cloning.md | | Fish Audio Voice Cloning API | C | 61.5 | 604 | voice.clone, voice.design, speech.tts | no | https://www.anchorterminal.com/tools/fish-audio-voice-cloning.md | | Gradium Voice Cloning and Voice Design | C | 60.8 | 637 | voice.clone, voice.design, speech.tts | no | https://www.anchorterminal.com/tools/gradium-voice-cloning.md | | Ming-3.0-tts-0.5B | C | 59.3 | open-weight #71 | speech.tts, voice.clone, voice.design | no | https://www.anchorterminal.com/tools/ming-3-0-tts-0-5b.md | | Inworld Realtime TTS-2 | C | 58.3 | 752 | speech.tts, voice.clone, voice.design | no | https://www.anchorterminal.com/tools/inworld-tts.md | | MiniMax Voice Clone and Voice Design | C | 55.1 | 880 | voice.clone, voice.design, speech.tts | no | https://www.anchorterminal.com/tools/minimax-voice-cloning.md | ## Panel reviews (0) Reviewed by the Anchor panel (https://www.anchorterminal.com/reviewers/index.md): . Desk reviews, written from public documentation, pricing, terms, source and status history on 1 October 2026. No calls made. For a desk review, the outcome says whether the reviewer's questions could be answered from public material: success, partial or failure. How reviews work: https://www.anchorterminal.com/reviews/how-it-works.md ## Notable - Inworld's pricing page lists instant voice cloning and voice design on every plan, including the free On-Demand plan with 100 custom voices, and no charge per clone (source: ) - The 30-second prompt limit applies to voices cloned on or after 15 August 2026. Older clones keep a 15-second prompt until re-cloned (source: ) - Zero Data Retention lists Voice Cloning audio samples as not eligible, and is an Enterprise add-on (source: ) - The MCP server in `@inworld/cli` 1.5.0 (29 October 2026) registers 40 tools, `clone_voice`, `design_voice` and `publish_voice` among them, and sets no readOnlyHint or destructiveHint on any (source: ) - The MCP `clone_voice` description still says about 5 seconds and 4 MB a sample, while the docs say 3 seconds and 12 MB (source: ) - status.inworld.ai is drawn from GitHub issues in inworld-ai/inworld-issue-status, with five components and none for cloning (source: ) - Professional Voice Cloning (10 minutes of audio or more, asynchronous training) is marked Beta and is not graded here (source: ) - #4 of 14 in Best voice cloning and custom voice APIs for AI agents: https://www.anchorterminal.com/best/voice-cloning/index.md - All 133 cloning comparisons: https://www.anchorterminal.com/compare/voice-cloning/index.md ## Compare - [Cartesia Voice Cloning API + MCP vs Inworld Instant Voice Cloning](https://www.anchorterminal.com/compare/cartesia-voice-cloning-vs-inworld-voice-cloning.md): C 58.3 vs C 61.9 - [ElevenLabs Voice Cloning and Voice Design API vs Inworld Instant Voice Cloning](https://www.anchorterminal.com/compare/elevenlabs-voice-cloning-vs-inworld-voice-cloning.md): BB 76.8 vs C 61.9 - [Fish Audio Voice Cloning API vs Inworld Instant Voice Cloning](https://www.anchorterminal.com/compare/fish-audio-voice-cloning-vs-inworld-voice-cloning.md): C 61.5 vs C 61.9 - [Gradium Voice Cloning and Voice Design vs Inworld Instant Voice Cloning](https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-inworld-voice-cloning.md): C 60.8 vs C 61.9 - [Hume Octave Voice Design and Cloning + MCP vs Inworld Instant Voice Cloning](https://www.anchorterminal.com/compare/hume-voice-cloning-vs-inworld-voice-cloning.md): C 54.6 vs C 61.9 - [Inworld Instant Voice Cloning vs MiniMax Voice Clone and Voice Design](https://www.anchorterminal.com/compare/inworld-voice-cloning-vs-minimax-voice-cloning.md): C 61.9 vs C 55.1 - [Inworld Instant Voice Cloning vs Murf Voice Cloning API](https://www.anchorterminal.com/compare/inworld-voice-cloning-vs-murf-voice-cloning.md): C 61.9 vs C 55.9 - [Inworld Instant Voice Cloning vs PlayHT Voice Cloning API](https://www.anchorterminal.com/compare/inworld-voice-cloning-vs-playht-voice-cloning.md): C 61.9 vs F 4.1 - [Inworld Instant Voice Cloning vs Resemble AI Voice Cloning API](https://www.anchorterminal.com/compare/inworld-voice-cloning-vs-resemble-ai-voice-cloning.md): C 61.9 vs D 52.5 - [Inworld Instant Voice Cloning vs Smallest.ai Instant Voice Clone](https://www.anchorterminal.com/compare/inworld-voice-cloning-vs-smallest-voice-clone.md): C 61.9 vs C 58 - [Inworld Instant Voice Cloning vs Soniox Voice Cloning](https://www.anchorterminal.com/compare/inworld-voice-cloning-vs-soniox-voice-cloning.md): C 61.9 vs B 68.7 - [Inworld Instant Voice Cloning vs Speechify API Voice Cloning](https://www.anchorterminal.com/compare/inworld-voice-cloning-vs-speechify-voice-cloning.md): C 61.9 vs BB 76.9 - [Inworld Instant Voice Cloning vs Ultravox Voice Cloning](https://www.anchorterminal.com/compare/inworld-voice-cloning-vs-ultravox-voice-cloning.md): C 61.9 vs E 40.9 - [Inworld Instant Voice Cloning vs VoiceStudio](https://www.anchorterminal.com/compare/inworld-voice-cloning-vs-voicestudio.md): C 61.9 vs D 50.4 ## Verify this listing For the vendor. The badge or a plain link to this page verifies the listing, from a page on inworld.ai or one of its subdomains, or the README at the root of github.com/inworld-ai/inworld-api-examples. It shows the listing is the vendor's and that the vendor knows it's here, and it never changes a grade, rank or review. The vendor sends the page's address to `POST https://www.anchorterminal.com/api/v1/verify` as `{"slug": "inworld-voice-cloning", "url": "…"}`, or calls the `verify_listing` tool at https://www.anchorterminal.com/mcp. We fetch the page once, then again every week; two failed checks in a row and the verification lapses, and a later pass restores it. What we check: https://www.anchorterminal.com/builders/index.md#verify HTML badge: ```html Inworld Instant Voice Cloning on Anchor Terminal ``` Markdown badge, for a README: ```markdown [![Inworld Instant Voice Cloning on Anchor Terminal](https://www.anchorterminal.com/badges/inworld-voice-cloning.svg)](https://www.anchorterminal.com/tools/inworld-voice-cloning) ``` Plain link: ```html Inworld Instant Voice Cloning on Anchor Terminal ``` ## Share this listing For the vendor. Sharing assets for social media, two PNGs of 1200 × 630 that say Inworld Instant Voice Cloning is listed on Anchor Terminal, with the vendor's logo and this page's address and no grade or score. - Dark: https://www.anchorterminal.com/assets/share/inworld-voice-cloning-dark.png - Light: https://www.anchorterminal.com/assets/share/inworld-voice-cloning-light.png --- Query this catalogue directly over MCP: https://www.anchorterminal.com/mcp (Streamable HTTP, no key). How: https://www.anchorterminal.com/agents/#mcp