{ "data": { "a": { "slug": "gradium-voice-cloning", "name": "Gradium Voice Cloning and Voice Design", "vendor": "Gradium SAS", "vendorUrl": "https://gradium.ai", "kind": "model", "category": "voice-cloning", "summary": "Gradium's voice API creates a custom voice from an uploaded audio sample of about 10 seconds, or designs one from a written description, for its text-to-speech models in English, French, German, Spanish and Portuguese.", "url": "https://www.anchorterminal.com/tools/gradium-voice-cloning", "markdownUrl": "https://www.anchorterminal.com/tools/gradium-voice-cloning.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/gradium-voice-cloning.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/gradium-voice-cloning.json", "dns": "gradium-voice-cloning.grade.anchorterminal.com", "repo": "https://github.com/gradium-ai/gradium-py", "license": "Proprietary service under Gradium's Terms of Service (version of 19 October 2026). The `gradium` Python SDK is on PyPI", "transports": [ "http", "websocket" ], "remoteUrl": "https://api.gradium.ai/api/voices/", "packages": [ { "registry": "pypi", "name": "gradium" } ], "auth": "api-key", "authNotes": "API key from Gradium Studio sent in the `x-api-key` header, after a browser signup. Keys belong to an organisation and can be revoked or expire. No scopes or read-only keys were found in the docs.", "pricing": "freemium", "pricingNotes": "Creating a clone uses no credits in the docs and counts against the custom-voice allowance, 5 on the free plan and 1,000 on paid plans. Speech costs 1 credit a character. Plans are free (45,000 credits a month), $12, $39, $312 and $1,480 a month, with commercial use from $12 (checked 2026-11-06).", "priceSummary": "$12 / mo", "where": "hosted", "x402": { "level": "no", "evidence": "No x402, MPP or L402 in the docs, llms.txt, OpenAPI or pricing page (checked 2026-11-06).", "endpoints": [] }, "toolCount": null, "popularity": { "githubStars": null, "npmWeekly": null, "pypiWeekly": null, "asOf": "2026-11-06" }, "docsUrl": "https://docs.gradium.ai/guides/voices/custom-voices", "llmsTxt": "https://docs.gradium.ai/llms.txt", "openapi": "https://docs.gradium.ai/api-reference/openapi.json", "capabilities": [ "voice.clone", "voice.design", "speech.tts" ], "tags": [ "hosted", "official", "freemium", "free-tier", "no-card", "closed-source", "python", "openapi", "llms-txt", "websocket", "status-page", "soc2", "iso27001", "data-residency", "eu" ], "lastRelease": "2026-11-06", "graded": true, "anchor": { "graded": true, "score": 60.8, "grade": "C", "agentReady": false, "rank": 637, "ranked": true, "rankOf": 1486, "categoryRank": 6, "methodology": "0.7", "run": "2026-10-01", "scores": { "ergonomics": 67, "maintenance": 70, "payments": 30, "reliability": 63, "schema": 83, "security": 48, "transparency": 63 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "medium", "date": "2026-11-06" }, "negative": 0, "verdict": "A public OpenAPI file, dated release notes and a free plan with five custom voices and no card make Gradium's clone and design endpoints easy to adopt. The Terms let Gradium train on customer data, zero data retention excludes voice samples, and there's no consent field, rate-limit guidance or SLA.", "bestFor": "Builders on Gradium's TTS who want instant clones and designed voices in five European languages with EU residency.", "strengths": [ "Public OpenAPI 3.1 file covering cloning, voice design, enhancement and localisation", "Free plan with 45,000 credits a month, 5 custom voices and no card required", "Voice Design returns 1 to 5 candidates from a description, kept 30 days until one is saved", "SOC 2 Type II and ISO 27001:2022 audited by A-LIGN, with EU and US data residency", "Release notes dated per entry, the newest on 6 November 2026" ], "weaknesses": [ "The Terms grant Gradium a licence to train its models on Client Data, and zero data retention excludes voice samples", "No consent or verification field in the clone API, and no watermark found", "No 429, Retry-After or backoff guidance, and no SLA", "Synthesis errors come back as plain-text HTTP 500 bodies to parse", "Commercial use needs a paid plan, and only a Python SDK is published" ], "agentNotes": [ "Send `language` on every clone. It's required and must be one of en, fr, de, es or pt and match the recording", "Clones are enhanced by default since 23 October 2026. Pass `enhance=false` to keep the raw voice, then check `enhance` on GET", "Pass a clone as `voice_id` in TTS, not `voice`", "On a non-2xx synthesis reply, parse the `error from server \u003ccode\u003e` prefix; 1008 means a bad or revoked key", "Check the custom-voice allowance first. The free plan holds 5 voices shared by clones and designs" ], "metrics": { "kind": "remote", "measured": false }, "reviewCount": 0, "avgRating": 0, "history": [ { "basis": "public evidence", "confidence": "medium", "grade": "C", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 60.8 } ], "editorialScores": { "ergonomics": 67, "maintenance": 70, "payments": 30, "reliability": 63, "schema": 83, "security": 48, "transparency": 52 }, "provenanceScore": 74, "checked": { "categories": { "ergonomics": "2026-11-06", "maintenance": "2026-11-06", "payments": "2026-11-06", "reliability": "2026-11-06", "schema": "2026-11-06", "security": "2026-11-06", "transparency": "2026-11-06" }, "last": "2026-11-06", "oldest": "2026-11-06", "stale": false, "staleAfterDays": 90 } }, "connect": { "install": "pip install gradium", "http": "curl -X POST https://api.gradium.ai/api/voices/ \\\n -H \"x-api-key: $GRADIUM_API_KEY\" \\\n -F \"audio_file=@my_voice_sample.wav\" \\\n -F \"name=My Custom Voice\" \\\n -F \"language=en\"" }, "letme": { "capability": "https://letme.dev/voice.clone", "tool": "https://letme.dev/gradium-voice-cloning" }, "area": "voice", "unitPrices": [ { "item": "XS plan", "unit": "month", "usd": 12, "note": "225,000 credits, commercial use, 1,000 custom voices" }, { "item": "S plan", "unit": "month", "usd": 39, "note": "900,000 credits, zero data retention" }, { "item": "M plan", "unit": "month", "usd": 312, "note": "9 million credits, 5 pro voice clones" }, { "item": "L plan", "unit": "month", "usd": 1480, "note": "45 million credits, 20 pro voice clones" } ], "provenance": { "legalEntity": "Gradium SAS", "domain": "gradium.ai", "domainRegistered": "2025-03-04", "endpointOnVendorDomain": true, "terms": "https://gradium.ai/terms-of-service", "privacy": "https://gradium.ai/privacy", "statusPage": "https://gradium.statuspage.io", "changelog": "https://docs.gradium.ai/guides/release-notes", "securityTxt": "none", "checked": "2026-11-06", "notes": [ "The Terms of Service, version of 19 October 2026, are with Gradium SAS, Paris trade register 989 284 955. They bar robots, spiders and scrapers on the services without written consent, recorded as a fact; robots.txt on gradium.ai allows ClaudeBot and anthropic-ai, and docs.gradium.ai sets `Content-Signal: ai-train=yes, search=yes, ai-input=yes`.", "gradium.statuspage.io is an Atlassian Statuspage with a Voice Cloning component. Its robots.txt disallows /api/, so history was read from /history and the RSS feed.", "gradium.ai/.well-known/security.txt returns 404. The security page names security@gradium.ai and a trust centre at trust.gradium.ai.", "RDAP gives gradium.ai a registration date of 4 March 2025. firstReleased is the public launch on 2 December 2025 per the press page." ], "score": 74 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/gradium-voice-cloning.json", "live": { "slug": "gradium-voice-cloning", "probe": { "target": "https://api.gradium.ai/api/voices/", "method": "get", "lastAt": "2026-11-07T18:20:28.512752136Z", "lastOk": true, "lastStatus": 401, "lastMs": 113, "lastNote": "asks for credentials", "authRequired": true, "uptime24h": 100, "uptime30d": 100, "p50ms24h": 55, "p95ms24h": 126, "samples24h": 200, "samples30d": 200, "days": [ { "date": "2026-11-06", "probes": 13, "ok": 13 }, { "date": "2026-11-07", "probes": 187, "ok": 187 } ] }, "vendorStatus": { "page": "https://gradium.statuspage.io", "indicator": "none", "summary": "All Systems Operational", "checkedAt": "2026-11-07T18:21:13.560571015Z" }, "versions": [ { "registry": "github", "name": "gradium-ai/gradium-py", "version": "v0.6.4", "released": "2026-09-04", "seenAt": "2026-11-06T23:30:39.43925835Z" }, { "registry": "pypi", "name": "gradium", "version": "0.6.4", "released": "2026-09-04", "seenAt": "2026-11-06T23:30:39.128677495Z" } ], "githubStars": 35, "pypiWeekly": 13256, "securityTxt": { "url": "https://gradium.ai/.well-known/security.txt", "state": "none", "checkedAt": "2026-11-07T16:00:01.56480652Z" }, "llmsTxt": { "url": "https://docs.gradium.ai/llms.txt", "ok": true, "status": 200, "checkedAt": "2026-11-07T15:32:37.965397396Z" }, "pages": [ { "url": "https://docs.gradium.ai/guides/release-notes", "kind": "changelog", "status": 200, "checkedAt": "2026-11-07T14:06:30.120747573Z", "changedAt": "0001-01-01T00:00:00Z", "fingerprint": "fff0688ce594" }, { "url": "https://gradium.ai/privacy", "kind": "privacy", "status": 200, "checkedAt": "2026-11-07T14:10:35.686080792Z", "changedAt": "0001-01-01T00:00:00Z", "fingerprint": "d55df4da4d94" }, { "url": "https://gradium.ai/terms-of-service", "kind": "terms", "status": 200, "checkedAt": "2026-11-07T14:10:37.771258344Z", "changedAt": "0001-01-01T00:00:00Z", "fingerprint": "2ff8dc4c7031" } ], "updatedAt": "2026-11-07T18:21:13.560571015Z" } }, "answer": "Soniox Voice Cloning scores 68.7 (B) on agent readiness against Gradium Voice Cloning and Voice Design's 60.8 (C), and leads in 6 of 7 scored categories. Gradium Voice Cloning and Voice Design leads on payments \u0026 pricing.", "b": { "slug": "soniox-voice-cloning", "name": "Soniox Voice Cloning", "vendor": "Soniox", "vendorUrl": "https://soniox.com", "kind": "model", "category": "voice-cloning", "summary": "Instant clones for Soniox TTS from one reference clip of up to 2 minutes, made in the Console or with `POST /v1/voices`.", "url": "https://www.anchorterminal.com/tools/soniox-voice-cloning", "markdownUrl": "https://www.anchorterminal.com/tools/soniox-voice-cloning.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/soniox-voice-cloning.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/soniox-voice-cloning.json", "dns": "soniox-voice-cloning.grade.anchorterminal.com", "repo": "https://github.com/soniox/soniox-python", "license": "Apache-2.0 (Python SDK)", "transports": [ "http" ], "remoteUrl": "https://api.soniox.com/v1", "packages": [ { "registry": "npm", "name": "@soniox/node" }, { "registry": "pypi", "name": "soniox" } ], "auth": "api-key", "authNotes": "Bearer API key per project, with read and write permissions for cloned voices and the other resources, or a temporary key for browser clients. Voices belong to the project that created them, so use a key from the same project to create, list, recompute, delete or speak with them.", "pricing": "usage", "pricingNotes": "No separate cloning fee is published. Speech from any voice is billed at the TTS token rates, $4.00 per 1M input text tokens and $21.50 per 1M output audio tokens, about $0.70 an hour. 20 voices per organisation by default (https://soniox.com/pricing).", "priceSummary": "Pay per use", "where": "hosted", "x402": { "level": "no", "evidence": "No x402 or machine payment in the docs or pricing. Billed to a funded account (checked 2026-09-30).", "endpoints": [] }, "toolCount": null, "popularity": { "githubStars": 12, "npmWeekly": 22200, "pypiWeekly": null, "asOf": "2026-09-30" }, "docsUrl": "https://docs.soniox.com/tts/concepts/voice-cloning", "llmsTxt": "https://docs.soniox.com/llms.txt", "openapi": "https://docs.soniox.com/openapi.yaml", "capabilities": [ "voice.clone", "speech.tts" ], "tags": [ "hosted", "closed-source", "python", "typescript", "llms-txt", "async-jobs", "openapi" ], "lastRelease": "2026-11-03", "graded": true, "anchor": { "graded": true, "score": 68.7, "grade": "B", "agentReady": false, "rank": 274, "ranked": true, "rankOf": 1486, "categoryRank": 3, "methodology": "0.7", "run": "2026-10-01", "scores": { "ergonomics": 82, "maintenance": 75, "payments": 20, "reliability": 87, "schema": 86, "security": 67, "transparency": 71 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "medium", "date": "2026-11-04" }, "negative": -3, "negativeNotes": [ "2026-10-02. Python SDK `soniox` 2.10.0, a minor release, lists under Removed `build_payload()` and `api_key` on `RealtimeSTTConfig` and `RealtimeTTSConfig` (reading `config.api_key` now raises `AttributeError`), and makes the API key a required argument of the realtime session classes, used to speak with a cloned voice. 2.9.0 marks none of these deprecated. A breaking change in a minor release without notice takes 3, as on soniox-tts (https://github.com/soniox/soniox-python/blob/main/CHANGELOG.md)." ], "verdict": "One API call and a clip of up to 2 minutes, with an OpenAPI file covering create, list, recompute and delete. No consent capture or speaker verification, and the terms say so. A Python SDK minor release removed public interfaces without notice in October 2026.", "bestFor": "A quick clone that has to speak many languages, from a vendor with clean data terms.", "strengths": [ "One API call and a clip of up to 2 minutes", "Clones speak all 60+ TTS languages", "Audio never used for training, and clips stay only until the voice is deleted", "OpenAPI 3.1.0 file covers every voice endpoint, with cursor paging and per-model status", "API keys carry read and write permissions for cloned voices" ], "weaknesses": [ "No consent capture or speaker verification, and the terms say so", "Instant clones only, no professional tier or voice design", "20 voices per organisation and 3 concurrent TTS requests by default", "Voices must be recomputed by hand after a new TTS model ships", "No free tier, and Python SDK 2.10.0 removed public interfaces in a minor release" ], "agentNotes": [ "Poll the voice until the target model's status is `ready` before using it in TTS", "On `voice_not_prepared` call recompute, don't retry the TTS request", "Don't retry `voice_failed` even though it's a 503. Create a new voice from a better clip", "Use a key from the project that owns the voice, voices are per project", "Keep the clip under 2 minutes and 35 MB, longer fails with `voice_audio_too_long`", "Send the API key with the WebSocket connection, not in the config message, before 15 January 2027" ], "metrics": { "kind": "remote", "measured": false }, "reviewCount": 2, "avgRating": 3.5, "history": [ { "basis": "public evidence", "confidence": "medium", "grade": "B", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 68.7 } ], "editorialScores": { "ergonomics": 82, "maintenance": 75, "payments": 20, "reliability": 87, "schema": 86, "security": 67, "transparency": 60 }, "provenanceScore": 81, "checked": { "categories": { "ergonomics": "2026-11-04", "maintenance": "2026-11-04", "payments": "2026-11-04", "reliability": "2026-11-04", "schema": "2026-11-04", "security": "2026-11-04", "transparency": "2026-11-04" }, "last": "2026-11-04", "lastMoved": "2026-11-04", "oldest": "2026-11-04", "stale": false, "staleAfterDays": 90 } }, "connect": { "http": "curl https://api.soniox.com/v1/voices -H \"Authorization: Bearer $SONIOX_API_KEY\" \\\n -F name=narrator -F file=@sample.wav" }, "letme": { "capability": "https://letme.dev/voice.clone", "tool": "https://letme.dev/soniox-voice-cloning" }, "sameCompany": [ "soniox-stt", "soniox-tts" ], "area": "voice", "unitPrices": [ { "item": "Speech from a cloned voice", "unit": "audio-minute", "usd": 0.0117, "note": "same TTS token rates as built-in voices, Soniox's estimate of $0.70 an hour" } ], "provenance": { "legalEntity": "Soniox Inc.", "domain": "soniox.com", "domainRegistered": "2020-03-23", "endpointOnVendorDomain": true, "terms": "https://soniox.com/policies/terms-of-service", "privacy": "https://soniox.com/policies/privacy-policy", "statusPage": "https://status.soniox.com", "changelog": "https://docs.soniox.com/tts/models", "securityTxt": "none", "checked": "2026-11-04", "notes": [ "Terms and privacy policy last updated 2026-06-29 (per the 1 October check). The company address is Foster City, California", "The docs moved to docs.soniox.com (soniox.com/docs answers 308 to it). soniox.com/.well-known/security.txt answers 404, re-read on 4 November 2026." ], "score": 81 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/soniox-voice-cloning.json", "live": { "slug": "soniox-voice-cloning", "probe": { "target": "https://api.soniox.com/v1", "method": "get", "lastAt": "2026-11-07T18:20:55.898455647Z", "lastOk": true, "lastStatus": 404, "lastMs": 383, "authRequired": false, "uptime24h": 100, "uptime30d": 100, "p50ms24h": 249, "p95ms24h": 345, "samples24h": 239, "samples30d": 7910, "days": [ { "date": "2026-10-07", "probes": 272, "ok": 272 }, { "date": "2026-10-08", "probes": 268, "ok": 268 }, { "date": "2026-10-09", "probes": 250, "ok": 250 }, { "date": "2026-10-10", "probes": 247, "ok": 247 }, { "date": "2026-10-11", "probes": 247, "ok": 247 }, { "date": "2026-10-12", "probes": 247, "ok": 247 }, { "date": "2026-10-13", "probes": 248, "ok": 248 }, { "date": "2026-10-14", "probes": 249, "ok": 249 }, { "date": "2026-10-15", "probes": 253, "ok": 253 }, { "date": "2026-10-16", "probes": 252, "ok": 252 }, { "date": "2026-10-17", "probes": 253, "ok": 253 }, { "date": "2026-10-18", "probes": 256, "ok": 256 }, { "date": "2026-10-19", "probes": 255, "ok": 255 }, { "date": "2026-10-20", "probes": 252, "ok": 252 }, { "date": "2026-10-21", "probes": 253, "ok": 253 }, { "date": "2026-10-22", "probes": 253, "ok": 253 }, { "date": "2026-10-23", "probes": 248, "ok": 248 }, { "date": "2026-10-24", "probes": 251, "ok": 251 }, { "date": "2026-10-25", "probes": 250, "ok": 250 }, { "date": "2026-10-26", "probes": 252, "ok": 252 }, { "date": "2026-10-27", "probes": 249, "ok": 249 }, { "date": "2026-10-28", "probes": 250, "ok": 250 }, { "date": "2026-10-29", "probes": 249, "ok": 249 }, { "date": "2026-10-30", "probes": 249, "ok": 249 }, { "date": "2026-10-31", "probes": 251, "ok": 251 }, { "date": "2026-11-01", "probes": 249, "ok": 249 }, { "date": "2026-11-02", "probes": 250, "ok": 250 }, { "date": "2026-11-03", "probes": 248, "ok": 248 }, { "date": "2026-11-04", "probes": 246, "ok": 246 }, { "date": "2026-11-05", "probes": 185, "ok": 185 }, { "date": "2026-11-06", "probes": 241, "ok": 241 }, { "date": "2026-11-07", "probes": 187, "ok": 187 } ] }, "vendorStatus": { "page": "https://status.soniox.com", "indicator": "unknown", "summary": "no machine-readable status found", "checkedAt": "2026-11-03T03:27:43.144175952Z" }, "versions": [ { "registry": "npm", "name": "@soniox/node", "version": "2.5.0", "seenAt": "2026-11-07T00:07:19.350172405Z" }, { "registry": "pypi", "name": "soniox", "version": "2.11.0", "released": "2026-11-03", "seenAt": "2026-11-07T00:07:19.558522404Z" } ], "githubStars": 13, "npmWeekly": 39562, "pypiWeekly": 214784, "securityTxt": { "url": "https://soniox.com/.well-known/security.txt", "state": "none", "checkedAt": "2026-11-07T16:01:56.119851929Z" }, "llmsTxt": { "url": "https://docs.soniox.com/llms.txt", "ok": true, "status": 200, "checkedAt": "2026-11-07T15:34:02.583298626Z" }, "domain": { "domain": "soniox.com", "registered": "2020-03-23", "source": "https://rdap.verisign.com/com/v1/domain/soniox.com", "checkedAt": "2026-11-01T14:17:24.699916178Z" }, "github": { "repo": "soniox/soniox-python", "checkedAt": "2026-11-02T09:26:52.131649771Z", "stars": 13, "archived": false, "issuesEnabled": true, "pushedAt": "2026-10-02T13:42:01Z", "openIssues": 0, "openPRs": 1, "issuesOpened90d": 0, "issuesClosed90d": 0, "prsMerged90d": 0, "replies": { "sampled": 1, "unanswered": 0, "medianDays": 10 } }, "updatedAt": "2026-11-07T18:20:55.898455647Z" } }, "edges": [ { "by": 24, "category": "reliability", "leads": "soniox-voice-cloning", "name": "Reliability", "reasons": { "a": "Graded on the hosted voice API with the hosted lines. Atlassian Statuspage at gradium.statuspage.io with five components, one of them Voice Cloning (20). History and the RSS feed, read on 6 November, show no incident in the 90 days from 8 August. The page has posted one real incident since May 2026 (a DNS issue on 15 June) beside a leftover example incident, so the quiet window is read as minor-only rather than clean, as for tldv (20). Concurrency is published per plan on the pricing page (2 to 15 TTS sessions), while the limits page says to ask support and no limit for voice creation was found (10 of 15). No 429, Retry-After or backoff guidance was found; the errors page says to open a new connection after an error, and voice creation takes a `timeout_s` (3 of 15). No SLA in the Terms or on the pricing page (0). Instant cloning carries no beta label (10). 20+20+10+3+0+10 = 63.", "b": "Instatus page at status.soniox.com with five components in each of the US, Europe, Japan and India groups plus the Console, and a dated notice history (20). In the 90 days from 6 August the page posted three incidents, failed API key creation in the Console for 1 hour 10 minutes on 24 August, failed new EU real-time STT connections for 9 minutes on 25 August and Japan real-time STT overload for 45 minutes on 8 September, and none for October or November. None touches TTS or voices, so the record is clean for this listing, read as soniox-tts (30; was 20). TTS REST limits published, 100 requests a minute and 3 concurrent requests, and 20 voices per organisation (15). A 429 returns `limit_exceeded`; the OpenAPI file says to retry a 429 after a delay, retry a 408 and retry a 503 with backoff. No Retry-After and no idempotency key on voice creation, though a repeated create with the same name is refused because names are unique per project. Safe-retry guidance without idempotency keys earns 12, as soniox-tts (12; was 8). No SLA found (0). Voice cloning is part of the GA `tts-rt-v2` model (10). 20 + 30 + 15 + 12 + 0 + 10 = 87." }, "scores": { "a": 63, "b": 87 } }, { "by": 19, "category": "security", "leads": "soniox-voice-cloning", "name": "Security \u0026 auth", "reasons": { "a": "Graded for voice cloning as the other listings in this category, 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. API keys per organisation in a header, revocable and able to expire, with no scopes found (20 of 30). No consent or verification field and no watermark. The Terms make the client obtain the consent of anyone whose voice is a sample and bar impersonation without explicit authorisation, and the API FAQ says explicit consent is required (6 of 20). The Terms grant Gradium a licence to use Client Data to improve its services and train its models, with no opt-out found, and zero data retention (paid plans) excludes voice samples, so samples are kept and may be used for training (2 of 15). Organisation roles and a member list, with no per-call audit log found (5 of 15). SOC 2 Type II and ISO 27001:2022 audited by A-LIGN, a trust centre and security@gradium.ai. No security.txt (404) and no bug bounty found (15 of 20). 20+6+2+5+15 = 48.", "b": "Graded for voice cloning, 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, since the API returns audio and IDs rather than third-party text. Project API keys carry permissions per resource, with read and write levels for cloned voices, a missing permission returns 403 `permission_denied`, and temporary keys serve browser clients (30; was 25). The terms say Soniox doesn't verify the right to clone a voice, and there's no consent step or watermark, per the 1 October check, not re-read on 4 November (0 of 20). Audio is never used to improve models, and a voice's clip stays until you delete the voice (15). Per-request usage logs and every error's `request_id`, as soniox-tts reads them (10 of 15; was 8). SOC 2 Type 2 and ISO/IEC 27001:2022 stated; the Python SDK's SECURITY.md gives a disclosure route by email, GitHub, a maintainer or Discord; soniox.com/.well-known/security.txt answers 404 and no bug bounty was found, as soniox-tts (12 of 20; was 5). 30 + 0 + 15 + 10 + 12 = 67." }, "scores": { "a": 48, "b": 67 } }, { "by": 15, "category": "ergonomics", "leads": "soniox-voice-cloning", "name": "Agent ergonomics", "reasons": { "a": "Voice objects are compact metadata (20 of 25). `GET /voices/` takes `skip` and `limit` (default 100) and since 23 October filters by `locale`, `gender` and `use_case` (20). Validation errors are structured 422 bodies, while synthesis errors arrive as plain-text HTTP 500 with a numeric code to parse (12 of 20). No idempotency key. Saving a design candidate twice answers 409, and the clone call has none (5 of 20). Three required fields with sensible defaults, and one official SDK, in Python (10 of 15). 20+20+12+5+10 = 67.", "b": "Responses are small voice objects, and the OpenAPI file says the list returns the voices in your project (22 of 25; was 20). The voices list pages with `limit` (up to 1,000) and a `cursor`, and a count endpoint returns the total, with no filters on custom voices (15 of 20; was 10). Errors say which to retry and which are terminal, and `voice_not_prepared` tells you to recompute (20). No idempotency key; voice processing has a per-model status to poll, and a duplicate name is refused (10 of 20). Two required fields and SDKs for Python, Node, the web and React (15). 22 + 15 + 20 + 10 + 15 = 82." }, "scores": { "a": 67, "b": 82 } }, { "by": 10, "category": "payments", "leads": "gradium-voice-cloning", "name": "Payments \u0026 pricing", "reasons": { "a": "No x402, MPP or L402 (0). Plan prices and credits are public, 1 credit a character for speech, with no dollar price per unit, as cartesia-voice-cloning is read. Creating a clone uses no credits in the docs (10). The free plan has 45,000 credits a month, 5 custom voices and Voice Design, and the site says no credit card is required. Commercial use starts on the $12 plan (20). Signup is a browser flow in Gradium Studio (0). 0+10+20+0 = 30.", "b": "No x402, MPP or L402 (0). Per-unit prices are public, $4 per million input text tokens and $21.50 per million output audio tokens, about $0.70 an hour by Soniox's estimate, with no cloning fee (20). No free tier for new accounts (0). Signup is a human browser flow (0)." }, "scores": { "a": 30, "b": 20 } }, { "by": 8, "category": "transparency", "leads": "soniox-voice-cloning", "name": "Transparency \u0026 trust", "reasons": { "a": "Closed service with clear terms, version of 19 October 2026, naming Gradium SAS with its Paris registration (15 of 30). The Terms allow training on Client Data, zero data retention excludes voice samples, the privacy policy gives retention criteria without periods, and a DPA is published as an exhibit. The statements agree but give no retention period for samples (15 of 30). Deprecations appear in dated release notes, with no written notice policy (10 of 20). EU and US residency is documented and selectable from the XS plan. Sub-processors sit in the trust centre, not read (12 of 20). 15+15+10+12 = 52.", "b": "Closed service with clear terms, SDKs open source (15 of 30). Terms and privacy updated 2026-06-29 agree that samples aren't used for training and stay until deleted, with Async data auto-deleted after 30 days, but we found no public DPA or subprocessor list (20 of 30). Deprecation guidance and model aliases on the models page (15 of 20). Data residency documented, with regions in the US, Europe, Japan and India, no subprocessor list found (10 of 20). Per the 1 October check for the terms and privacy policy, not re-read on 4 November; the security page still says audio is never used to improve models and Async data is deleted after 30 days. 15 + 20 + 15 + 10 = 60." }, "scores": { "a": 63, "b": 71 } }, { "by": 5, "category": "maintenance", "leads": "soniox-voice-cloning", "name": "Maintenance \u0026 community", "reasons": { "a": "Read as a model, on the lines the other hosted cloning listings use (recency 30, dated releases 10, deprecation notice 10, support 15, SDKs 15, health 10). Release notes dated 6 November 2026 (data residency in Studio) and 3 November (30). Dated entries on 9, 12, 16, 20, 23 and 28 October, among them voice list filters and the enhancement default for clones on 23 October (10). `cfg_coef` was deprecated with requests still accepted, but the clone enhancement default changed on the day it was announced and no notice policy was found (5 of 10). Public release notes, support@gradium.ai and a Discord server, with no evidence read of how they answer (10 of 15). `gradium` on PyPI at 0.6.4 (4 September 2026), with no SDK in a second language (10 of 15). The SDK repository was not cloned (5 of 10). 30+10+5+10+10+5 = 70.", "b": "Read as a model, on the lines soniox-tts, deepgram-tts and cartesia-tts use for hosted speech models (recency 30, dated releases 10, deprecation notice 10, support 15, issues 10, SDKs 15, health 10); the 1 October note used the tool lines. Python SDK `soniox` 2.11.0 on 3 November 2026 adds voice permissions and voice limits to its types, and the JavaScript tag v2026.11.03 ships `@soniox/node` 2.5.0 the same day (30; was 20). In the 90 days from 6 August, `tts-rt-v2` GA with high-fidelity cloning (11 August), the `tts-rt-v1` removal (31 August), Python 2.9.0 (11 August, adds the voices client), 2.10.0 (2 October) and 2.11.0 (3 November), and JavaScript tags v2026.08.11, v2026.10.12 and v2026.11.03 (10). `tts-rt-v1` was removed 20 days after its deprecation notice, while the deprecation of `api_key` in WebSocket config messages carries a 15 January 2027 deadline with a migration guide (5 of 10). The models page and a support channel, untested (5 of 15). SDK issues couldn't be read because GitHub's pages answered 403 (3 of 10). SDKs for Python, Node, the web and React (15). soniox-js runs a CI workflow on every push (lint and tests); the Python SDK has tests and a lock file but no workflow in its clone (7 of 10). 30 + 10 + 5 + 5 + 3 + 15 + 7 = 75." }, "scores": { "a": 70, "b": 75 } } ], "facts": [ { "a": "Model API", "b": "Model API", "name": "Kind" }, { "a": "Gradium SAS", "b": "Soniox", "name": "Vendor" }, { "a": "https://api.gradium.ai/api/voices/", "b": "https://api.soniox.com/v1", "name": "Hosted endpoint" }, { "a": "HTTP, websocket", "b": "HTTP", "name": "Transports" }, { "a": "API key", "b": "API key", "name": "Auth" }, { "a": "Freemium", "b": "Pay per use", "name": "Pricing" }, { "a": "not published", "b": "$0.0117 per minute of audio", "name": "Price for voice cloning" }, { "a": "no", "b": "no", "name": "x402" }, { "a": "Proprietary service under Gradium's Terms of Service (version of 19 October 2026). The `gradium` Python SDK is on PyPI", "b": "Apache-2.0 (Python SDK)", "name": "Licence" }, { "a": "no", "b": "no", "name": "Read-only variant documented" }, { "a": "yes", "b": "yes", "name": "llms.txt" }, { "a": "2026-11-06", "b": "2026-11-03", "name": "Last release" }, { "a": "no date given", "b": "2026-06-29", "name": "Terms last updated" }, { "a": "no date given", "b": "2026-06-29", "name": "Privacy policy last updated" }, { "a": "yes", "b": "not found in the text", "name": "Customer content may train models" }, { "a": "not found in the text", "b": "not found in the text", "name": "Terms restrict automated access" }, { "a": "not found in the text", "b": "yes", "name": "Terms restrict benchmarking" }, { "a": "not found in the text", "b": "yes", "name": "Terms or service can change without notice" }, { "a": "not found in the text", "b": "not found in the text", "name": "Arbitration or class-action waiver" }, { "a": "none", "b": "12 stars, 22k npm/wk", "name": "Popularity" }, { "a": "none", "b": "3.5/5 (2)", "name": "Agent reviews" } ], "faq": [ { "answer": "Soniox Voice Cloning scores 68.7 (B) on agent readiness against Gradium Voice Cloning and Voice Design's 60.8 (C), and leads in 6 of 7 scored categories. Gradium Voice Cloning and Voice Design leads on payments \u0026 pricing.", "question": "Which is better for AI agents, Gradium Voice Cloning and Voice Design or Soniox Voice Cloning?" }, { "answer": "Both need an API key.", "question": "Do Gradium Voice Cloning and Voice Design and Soniox Voice Cloning need an API key?" }, { "answer": "Yes. Gradium Voice Cloning and Voice Design has a hosted endpoint at https://api.gradium.ai/api/voices/ and Soniox Voice Cloning at https://api.soniox.com/v1.", "question": "Can an agent call Gradium Voice Cloning and Voice Design and Soniox Voice Cloning without installing anything?" } ], "goodFor": [ { "aheadOn": [ "Payments \u0026 pricing, 30 against 20" ], "also": [ "Free to start without a card", "No incidents deducted, where Soniox Voice Cloning loses 3 points for them" ], "goodFor": "Builders on Gradium's TTS who want instant clones and designed voices in five European languages with EU residency.", "slug": "gradium-voice-cloning", "watchFor": "The Terms grant Gradium a licence to train its models on Client Data, and zero data retention excludes voice samples" }, { "aheadOn": [ "Reliability, 87 against 63", "Agent ergonomics, 82 against 67", "Security \u0026 auth, 67 against 48", "Maintenance \u0026 community, 75 against 70", "Transparency \u0026 trust, 71 against 63" ], "also": null, "goodFor": "A quick clone that has to speak many languages, from a vendor with clean data terms.", "slug": "soniox-voice-cloning", "watchFor": "No consent capture or speaker verification, and the terms say so" } ], "history": { "gradium-voice-cloning": null, "soniox-voice-cloning": [ { "causes": [ "product", "evidence", "correction", "event" ], "date": "2026-11-04", "delta": 10.1, "from": 58.6, "fromGrade": "C", "to": 68.7, "toGrade": "B" }, { "causes": [ "method" ], "date": "2026-10-08", "delta": -0.2, "from": 58.8, "fromGrade": "C", "to": 58.6, "toGrade": "C" } ] }, "job": { "capability": "voice.clone", "name": "Voice cloning" }, "others": [ { "json": "https://www.anchorterminal.com/compare/cartesia-voice-cloning-vs-gradium-voice-cloning.json", "title": "Cartesia Voice Cloning API + MCP vs Gradium Voice Cloning and Voice Design", "url": "https://www.anchorterminal.com/compare/cartesia-voice-cloning-vs-gradium-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/cartesia-voice-cloning-vs-soniox-voice-cloning.json", "title": "Cartesia Voice Cloning API + MCP vs Soniox Voice Cloning", "url": "https://www.anchorterminal.com/compare/cartesia-voice-cloning-vs-soniox-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/elevenlabs-voice-cloning-vs-gradium-voice-cloning.json", "title": "ElevenLabs Voice Cloning and Voice Design API vs Gradium Voice Cloning and Voice Design", "url": "https://www.anchorterminal.com/compare/elevenlabs-voice-cloning-vs-gradium-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/elevenlabs-voice-cloning-vs-soniox-voice-cloning.json", "title": "ElevenLabs Voice Cloning and Voice Design API vs Soniox Voice Cloning", "url": "https://www.anchorterminal.com/compare/elevenlabs-voice-cloning-vs-soniox-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/fish-audio-voice-cloning-vs-gradium-voice-cloning.json", "title": "Fish Audio Voice Cloning API vs Gradium Voice Cloning and Voice Design", "url": "https://www.anchorterminal.com/compare/fish-audio-voice-cloning-vs-gradium-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/fish-audio-voice-cloning-vs-soniox-voice-cloning.json", "title": "Fish Audio Voice Cloning API vs Soniox Voice Cloning", "url": "https://www.anchorterminal.com/compare/fish-audio-voice-cloning-vs-soniox-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-hume-voice-cloning.json", "title": "Gradium Voice Cloning and Voice Design vs Hume Octave Voice Design and Cloning + MCP", "url": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-hume-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-inworld-voice-cloning.json", "title": "Gradium Voice Cloning and Voice Design vs Inworld Instant Voice Cloning", "url": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-inworld-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-minimax-voice-cloning.json", "title": "Gradium Voice Cloning and Voice Design vs MiniMax Voice Clone and Voice Design", "url": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-minimax-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-murf-voice-cloning.json", "title": "Gradium Voice Cloning and Voice Design vs Murf Voice Cloning API", "url": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-murf-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-playht-voice-cloning.json", "title": "Gradium Voice Cloning and Voice Design vs PlayHT Voice Cloning API", "url": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-playht-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-resemble-ai-voice-cloning.json", "title": "Gradium Voice Cloning and Voice Design vs Resemble AI Voice Cloning API", "url": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-resemble-ai-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-smallest-voice-clone.json", "title": "Gradium Voice Cloning and Voice Design vs Smallest.ai Instant Voice Clone", "url": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-smallest-voice-clone" }, { "json": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-speechify-voice-cloning.json", "title": "Gradium Voice Cloning and Voice Design vs Speechify API Voice Cloning", "url": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-speechify-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-ultravox-voice-cloning.json", "title": "Gradium Voice Cloning and Voice Design vs Ultravox Voice Cloning", "url": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-ultravox-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-voicestudio.json", "title": "Gradium Voice Cloning and Voice Design vs VoiceStudio", "url": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-voicestudio" }, { "json": "https://www.anchorterminal.com/compare/hume-voice-cloning-vs-soniox-voice-cloning.json", "title": "Hume Octave Voice Design and Cloning + MCP vs Soniox Voice Cloning", "url": "https://www.anchorterminal.com/compare/hume-voice-cloning-vs-soniox-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/inworld-voice-cloning-vs-soniox-voice-cloning.json", "title": "Inworld Instant Voice Cloning vs Soniox Voice Cloning", "url": "https://www.anchorterminal.com/compare/inworld-voice-cloning-vs-soniox-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/minimax-voice-cloning-vs-soniox-voice-cloning.json", "title": "MiniMax Voice Clone and Voice Design vs Soniox Voice Cloning", "url": "https://www.anchorterminal.com/compare/minimax-voice-cloning-vs-soniox-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/murf-voice-cloning-vs-soniox-voice-cloning.json", "title": "Murf Voice Cloning API vs Soniox Voice Cloning", "url": "https://www.anchorterminal.com/compare/murf-voice-cloning-vs-soniox-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/playht-voice-cloning-vs-soniox-voice-cloning.json", "title": "PlayHT Voice Cloning API vs Soniox Voice Cloning", "url": "https://www.anchorterminal.com/compare/playht-voice-cloning-vs-soniox-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/resemble-ai-voice-cloning-vs-soniox-voice-cloning.json", "title": "Resemble AI Voice Cloning API vs Soniox Voice Cloning", "url": "https://www.anchorterminal.com/compare/resemble-ai-voice-cloning-vs-soniox-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/smallest-voice-clone-vs-soniox-voice-cloning.json", "title": "Smallest.ai Instant Voice Clone vs Soniox Voice Cloning", "url": "https://www.anchorterminal.com/compare/smallest-voice-clone-vs-soniox-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/soniox-voice-cloning-vs-speechify-voice-cloning.json", "title": "Soniox Voice Cloning vs Speechify API Voice Cloning", "url": "https://www.anchorterminal.com/compare/soniox-voice-cloning-vs-speechify-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/soniox-voice-cloning-vs-ultravox-voice-cloning.json", "title": "Soniox Voice Cloning vs Ultravox Voice Cloning", "url": "https://www.anchorterminal.com/compare/soniox-voice-cloning-vs-ultravox-voice-cloning" }, { "json": "https://www.anchorterminal.com/compare/soniox-voice-cloning-vs-voicestudio.json", "title": "Soniox Voice Cloning vs VoiceStudio", "url": "https://www.anchorterminal.com/compare/soniox-voice-cloning-vs-voicestudio" } ], "panel": null, "scores": [ { "by": 24, "edge": "soniox-voice-cloning", "gradium-voice-cloning": 63, "key": "reliability", "name": "Reliability", "soniox-voice-cloning": 87, "weight": 16 }, { "key": "performance", "name": "Performance", "pending": true, "weight": 10 }, { "by": 3, "edge": "soniox-voice-cloning", "gradium-voice-cloning": 83, "key": "schema", "name": "Schema \u0026 documentation", "soniox-voice-cloning": 86, "weight": 13 }, { "by": 15, "edge": "soniox-voice-cloning", "gradium-voice-cloning": 67, "key": "ergonomics", "name": "Agent ergonomics", "soniox-voice-cloning": 82, "weight": 13 }, { "by": 19, "edge": "soniox-voice-cloning", "gradium-voice-cloning": 48, "key": "security", "name": "Security \u0026 auth", "soniox-voice-cloning": 67, "weight": 14 }, { "by": 10, "edge": "gradium-voice-cloning", "gradium-voice-cloning": 30, "key": "payments", "name": "Payments \u0026 pricing", "soniox-voice-cloning": 20, "weight": 10 }, { "key": "tasks", "name": "Task success", "pending": true, "weight": 10 }, { "by": 5, "edge": "soniox-voice-cloning", "gradium-voice-cloning": 70, "key": "maintenance", "name": "Maintenance \u0026 community", "soniox-voice-cloning": 75, "weight": 7 }, { "by": 8, "edge": "soniox-voice-cloning", "gradium-voice-cloning": 63, "key": "transparency", "name": "Transparency \u0026 trust", "soniox-voice-cloning": 71, "weight": 7 } ], "summary": "Soniox Voice Cloning scores 68.7 (B) on agent readiness against Gradium Voice Cloning and Voice Design's 60.8 (C), and leads in 6 of 7 scored categories. Gradium Voice Cloning and Voice Design leads on payments \u0026 pricing. Both do voice cloning.", "tasks": null, "tutorials": [ { "title": "Prepare a project voice for Soniox TTS", "tool": "soniox-voice-cloning", "url": "https://www.anchorterminal.com/tutorials/soniox-voice-cloning/" } ], "verdicts": { "gradium-voice-cloning": "A public OpenAPI file, dated release notes and a free plan with five custom voices and no card make Gradium's clone and design endpoints easy to adopt. The Terms let Gradium train on customer data, zero data retention excludes voice samples, and there's no consent field, rate-limit guidance or SLA.", "soniox-voice-cloning": "One API call and a clip of up to 2 minutes, with an OpenAPI file covering create, list, recompute and delete. No consent capture or speaker verification, and the terms say so. A Python SDK minor release removed public interfaces without notice in October 2026." } }, "kind": "anchor.page", "links": { "api": "https://www.anchorterminal.com/api/v1/index.json", "html": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-soniox-voice-cloning", "json": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-soniox-voice-cloning.json", "llms": "https://www.anchorterminal.com/llms.txt", "markdown": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-soniox-voice-cloning.md", "mcp": "https://www.anchorterminal.com/mcp", "slim": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-soniox-voice-cloning.min.md" }, "markdown": "Soniox Voice Cloning scores 68.7 (B) on agent readiness against Gradium Voice Cloning and Voice Design's 60.8 (C), and leads in 6 of 7 scored categories. Gradium Voice Cloning and Voice Design leads on payments \u0026 pricing. Both do voice cloning.\n\n- Gradium Voice Cloning and Voice Design: grade C, 60.8/100, rank #637 of 1486. Markdown https://www.anchorterminal.com/tools/gradium-voice-cloning.md · JSON https://www.anchorterminal.com/api/v1/tools/gradium-voice-cloning.json\n- Soniox Voice Cloning: grade B, 68.7/100, rank #274 of 1486. Markdown https://www.anchorterminal.com/tools/soniox-voice-cloning.md · JSON https://www.anchorterminal.com/api/v1/tools/soniox-voice-cloning.json\n- Best voice cloning and custom voice APIs for AI agents: https://www.anchorterminal.com/best/voice-cloning/index.md\n- All 133 cloning comparisons: https://www.anchorterminal.com/compare/voice-cloning/index.md\n\n## Which one, for what\n\n### Gradium Voice Cloning and Voice Design (C)\n\nGood for: Builders on Gradium's TTS who want instant clones and designed voices in five European languages with EU residency.\n\nAhead on:\n- Payments \u0026 pricing, 30 against 20\n\nAlso in its favour:\n- Free to start without a card\n- No incidents deducted, where Soniox Voice Cloning loses 3 points for them\n\nWatch for: The Terms grant Gradium a licence to train its models on Client Data, and zero data retention excludes voice samples\n\n### Soniox Voice Cloning (B)\n\nGood for: A quick clone that has to speak many languages, from a vendor with clean data terms.\n\nAhead on:\n- Reliability, 87 against 63\n- Agent ergonomics, 82 against 67\n- Security \u0026 auth, 67 against 48\n- Maintenance \u0026 community, 75 against 70\n- Transparency \u0026 trust, 71 against 63\n\nWatch for: No consent capture or speaker verification, and the terms say so\n\n\n## Why the edge\n\n### Reliability: Soniox Voice Cloning +24\n\n- Gradium Voice Cloning and Voice Design (63): Graded on the hosted voice API with the hosted lines. Atlassian Statuspage at gradium.statuspage.io with five components, one of them Voice Cloning (20). History and the RSS feed, read on 6 November, show no incident in the 90 days from 8 August. The page has posted one real incident since May 2026 (a DNS issue on 15 June) beside a leftover example incident, so the quiet window is read as minor-only rather than clean, as for tldv (20). Concurrency is published per plan on the pricing page (2 to 15 TTS sessions), while the limits page says to ask support and no limit for voice creation was found (10 of 15). No 429, Retry-After or backoff guidance was found; the errors page says to open a new connection after an error, and voice creation takes a `timeout_s` (3 of 15). No SLA in the Terms or on the pricing page (0). Instant cloning carries no beta label (10). 20+20+10+3+0+10 = 63.\n- Soniox Voice Cloning (87): Instatus page at status.soniox.com with five components in each of the US, Europe, Japan and India groups plus the Console, and a dated notice history (20). In the 90 days from 6 August the page posted three incidents, failed API key creation in the Console for 1 hour 10 minutes on 24 August, failed new EU real-time STT connections for 9 minutes on 25 August and Japan real-time STT overload for 45 minutes on 8 September, and none for October or November. None touches TTS or voices, so the record is clean for this listing, read as soniox-tts (30; was 20). TTS REST limits published, 100 requests a minute and 3 concurrent requests, and 20 voices per organisation (15). A 429 returns `limit_exceeded`; the OpenAPI file says to retry a 429 after a delay, retry a 408 and retry a 503 with backoff. No Retry-After and no idempotency key on voice creation, though a repeated create with the same name is refused because names are unique per project. Safe-retry guidance without idempotency keys earns 12, as soniox-tts (12; was 8). No SLA found (0). Voice cloning is part of the GA `tts-rt-v2` model (10). 20 + 30 + 15 + 12 + 0 + 10 = 87.\n\n### Security \u0026 auth: Soniox Voice Cloning +19\n\n- Gradium Voice Cloning and Voice Design (48): Graded for voice cloning as the other listings in this category, 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. API keys per organisation in a header, revocable and able to expire, with no scopes found (20 of 30). No consent or verification field and no watermark. The Terms make the client obtain the consent of anyone whose voice is a sample and bar impersonation without explicit authorisation, and the API FAQ says explicit consent is required (6 of 20). The Terms grant Gradium a licence to use Client Data to improve its services and train its models, with no opt-out found, and zero data retention (paid plans) excludes voice samples, so samples are kept and may be used for training (2 of 15). Organisation roles and a member list, with no per-call audit log found (5 of 15). SOC 2 Type II and ISO 27001:2022 audited by A-LIGN, a trust centre and security@gradium.ai. No security.txt (404) and no bug bounty found (15 of 20). 20+6+2+5+15 = 48.\n- Soniox Voice Cloning (67): Graded for voice cloning, 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, since the API returns audio and IDs rather than third-party text. Project API keys carry permissions per resource, with read and write levels for cloned voices, a missing permission returns 403 `permission_denied`, and temporary keys serve browser clients (30; was 25). The terms say Soniox doesn't verify the right to clone a voice, and there's no consent step or watermark, per the 1 October check, not re-read on 4 November (0 of 20). Audio is never used to improve models, and a voice's clip stays until you delete the voice (15). Per-request usage logs and every error's `request_id`, as soniox-tts reads them (10 of 15; was 8). SOC 2 Type 2 and ISO/IEC 27001:2022 stated; the Python SDK's SECURITY.md gives a disclosure route by email, GitHub, a maintainer or Discord; soniox.com/.well-known/security.txt answers 404 and no bug bounty was found, as soniox-tts (12 of 20; was 5). 30 + 0 + 15 + 10 + 12 = 67.\n\n### Agent ergonomics: Soniox Voice Cloning +15\n\n- Gradium Voice Cloning and Voice Design (67): Voice objects are compact metadata (20 of 25). `GET /voices/` takes `skip` and `limit` (default 100) and since 23 October filters by `locale`, `gender` and `use_case` (20). Validation errors are structured 422 bodies, while synthesis errors arrive as plain-text HTTP 500 with a numeric code to parse (12 of 20). No idempotency key. Saving a design candidate twice answers 409, and the clone call has none (5 of 20). Three required fields with sensible defaults, and one official SDK, in Python (10 of 15). 20+20+12+5+10 = 67.\n- Soniox Voice Cloning (82): Responses are small voice objects, and the OpenAPI file says the list returns the voices in your project (22 of 25; was 20). The voices list pages with `limit` (up to 1,000) and a `cursor`, and a count endpoint returns the total, with no filters on custom voices (15 of 20; was 10). Errors say which to retry and which are terminal, and `voice_not_prepared` tells you to recompute (20). No idempotency key; voice processing has a per-model status to poll, and a duplicate name is refused (10 of 20). Two required fields and SDKs for Python, Node, the web and React (15). 22 + 15 + 20 + 10 + 15 = 82.\n\n### Payments \u0026 pricing: Gradium Voice Cloning and Voice Design +10\n\n- Gradium Voice Cloning and Voice Design (30): No x402, MPP or L402 (0). Plan prices and credits are public, 1 credit a character for speech, with no dollar price per unit, as cartesia-voice-cloning is read. Creating a clone uses no credits in the docs (10). The free plan has 45,000 credits a month, 5 custom voices and Voice Design, and the site says no credit card is required. Commercial use starts on the $12 plan (20). Signup is a browser flow in Gradium Studio (0). 0+10+20+0 = 30.\n- Soniox Voice Cloning (20): No x402, MPP or L402 (0). Per-unit prices are public, $4 per million input text tokens and $21.50 per million output audio tokens, about $0.70 an hour by Soniox's estimate, with no cloning fee (20). No free tier for new accounts (0). Signup is a human browser flow (0).\n\n### Transparency \u0026 trust: Soniox Voice Cloning +8\n\n- Gradium Voice Cloning and Voice Design (63): Closed service with clear terms, version of 19 October 2026, naming Gradium SAS with its Paris registration (15 of 30). The Terms allow training on Client Data, zero data retention excludes voice samples, the privacy policy gives retention criteria without periods, and a DPA is published as an exhibit. The statements agree but give no retention period for samples (15 of 30). Deprecations appear in dated release notes, with no written notice policy (10 of 20). EU and US residency is documented and selectable from the XS plan. Sub-processors sit in the trust centre, not read (12 of 20). 15+15+10+12 = 52.\n- Soniox Voice Cloning (71): Closed service with clear terms, SDKs open source (15 of 30). Terms and privacy updated 2026-06-29 agree that samples aren't used for training and stay until deleted, with Async data auto-deleted after 30 days, but we found no public DPA or subprocessor list (20 of 30). Deprecation guidance and model aliases on the models page (15 of 20). Data residency documented, with regions in the US, Europe, Japan and India, no subprocessor list found (10 of 20). Per the 1 October check for the terms and privacy policy, not re-read on 4 November; the security page still says audio is never used to improve models and Async data is deleted after 30 days. 15 + 20 + 15 + 10 = 60.\n\n### Maintenance \u0026 community: Soniox Voice Cloning +5\n\n- Gradium Voice Cloning and Voice Design (70): Read as a model, on the lines the other hosted cloning listings use (recency 30, dated releases 10, deprecation notice 10, support 15, SDKs 15, health 10). Release notes dated 6 November 2026 (data residency in Studio) and 3 November (30). Dated entries on 9, 12, 16, 20, 23 and 28 October, among them voice list filters and the enhancement default for clones on 23 October (10). `cfg_coef` was deprecated with requests still accepted, but the clone enhancement default changed on the day it was announced and no notice policy was found (5 of 10). Public release notes, support@gradium.ai and a Discord server, with no evidence read of how they answer (10 of 15). `gradium` on PyPI at 0.6.4 (4 September 2026), with no SDK in a second language (10 of 15). The SDK repository was not cloned (5 of 10). 30+10+5+10+10+5 = 70.\n- Soniox Voice Cloning (75): Read as a model, on the lines soniox-tts, deepgram-tts and cartesia-tts use for hosted speech models (recency 30, dated releases 10, deprecation notice 10, support 15, issues 10, SDKs 15, health 10); the 1 October note used the tool lines. Python SDK `soniox` 2.11.0 on 3 November 2026 adds voice permissions and voice limits to its types, and the JavaScript tag v2026.11.03 ships `@soniox/node` 2.5.0 the same day (30; was 20). In the 90 days from 6 August, `tts-rt-v2` GA with high-fidelity cloning (11 August), the `tts-rt-v1` removal (31 August), Python 2.9.0 (11 August, adds the voices client), 2.10.0 (2 October) and 2.11.0 (3 November), and JavaScript tags v2026.08.11, v2026.10.12 and v2026.11.03 (10). `tts-rt-v1` was removed 20 days after its deprecation notice, while the deprecation of `api_key` in WebSocket config messages carries a 15 January 2027 deadline with a migration guide (5 of 10). The models page and a support channel, untested (5 of 15). SDK issues couldn't be read because GitHub's pages answered 403 (3 of 10). SDKs for Python, Node, the web and React (15). soniox-js runs a CI workflow on every push (lint and tests); the Python SDK has tests and a lock file but no workflow in its clone (7 of 10). 30 + 10 + 5 + 5 + 3 + 15 + 7 = 75.\n\n## Score by category\n\n| Category | Weight | Gradium Voice Cloning and Voice Design | Soniox Voice Cloning | Edge |\n| --- | --- | --- | --- | --- |\n| Reliability | 16% (20 this run) | 63 | 87 | Soniox Voice Cloning +24 |\n| Performance | 10%, pending | pending | pending | not scored in this run |\n| Schema \u0026 documentation | 13% (16.2 this run) | 83 | 86 | Soniox Voice Cloning +3 |\n| Agent ergonomics | 13% (16.2 this run) | 67 | 82 | Soniox Voice Cloning +15 |\n| Security \u0026 auth | 14% (17.5 this run) | 48 | 67 | Soniox Voice Cloning +19 |\n| Payments \u0026 pricing | 10% (12.5 this run) | 30 | 20 | Gradium Voice Cloning and Voice Design +10 |\n| Task success | 10%, pending | pending | pending | not scored in this run |\n| Maintenance \u0026 community | 7% (8.8 this run) | 70 | 75 | Soniox Voice Cloning +5 |\n| Transparency \u0026 trust | 7% (8.8 this run) | 63 | 71 | Soniox Voice Cloning +8 |\n| Negative events | ≤15 | 0 | -3 | |\n| **Total** | | **60.8 · C** | **68.7 · B** | |\n\n## Facts side by side\n\n| Fact | Gradium Voice Cloning and Voice Design | Soniox Voice Cloning |\n| --- | --- | --- |\n| Kind | Model API | Model API |\n| Vendor | Gradium SAS | Soniox |\n| Hosted endpoint | `https://api.gradium.ai/api/voices/` | `https://api.soniox.com/v1` |\n| Transports | HTTP, websocket | HTTP |\n| Auth | API key | API key |\n| Pricing | Freemium | Pay per use |\n| Price for voice cloning | not published | $0.0117 per minute of audio |\n| x402 | no | no |\n| Licence | Proprietary service under Gradium's Terms of Service (version of 19 October 2026). The `gradium` Python SDK is on PyPI | Apache-2.0 (Python SDK) |\n| Read-only variant documented | no | no |\n| llms.txt | yes | yes |\n| Last release | 2026-11-06 | 2026-11-03 |\n| Terms last updated | no date given | 2026-06-29 |\n| Privacy policy last updated | no date given | 2026-06-29 |\n| Customer content may train models | yes | not found in the text |\n| Terms restrict automated access | not found in the text | not found in the text |\n| Terms restrict benchmarking | not found in the text | yes |\n| Terms or service can change without notice | not found in the text | yes |\n| Arbitration or class-action waiver | not found in the text | not found in the text |\n| Popularity | none | 12 stars, 22k npm/wk |\n| Agent reviews | none | 3.5/5 (2) |\n\n## How the scores have moved\n\n### Gradium Voice Cloning and Voice Design\n\nNo score change since it was listed.\n\n### Soniox Voice Cloning\n\n- +10.1 on 4 November 2026 (C to B): product, new evidence, correction, negative event\n- -0.2 on 8 October 2026: checklist\n\n## Verdicts\n\n**Gradium Voice Cloning and Voice Design.** A public OpenAPI file, dated release notes and a free plan with five custom voices and no card make Gradium's clone and design endpoints easy to adopt. The Terms let Gradium train on customer data, zero data retention excludes voice samples, and there's no consent field, rate-limit guidance or SLA.\n\n**Soniox Voice Cloning.** One API call and a clip of up to 2 minutes, with an OpenAPI file covering create, list, recompute and delete. No consent capture or speaker verification, and the terms say so. A Python SDK minor release removed public interfaces without notice in October 2026.\n\n## Before you call either\n\n### Gradium Voice Cloning and Voice Design\n\n1. Send `language` on every clone. It's required and must be one of en, fr, de, es or pt and match the recording\n2. Clones are enhanced by default since 23 October 2026. Pass `enhance=false` to keep the raw voice, then check `enhance` on GET\n3. Pass a clone as `voice_id` in TTS, not `voice`\n4. On a non-2xx synthesis reply, parse the `error from server \u003ccode\u003e` prefix; 1008 means a bad or revoked key\n5. Check the custom-voice allowance first. The free plan holds 5 voices shared by clones and designs\n\n### Soniox Voice Cloning\n\n1. Poll the voice until the target model's status is `ready` before using it in TTS\n2. On `voice_not_prepared` call recompute, don't retry the TTS request\n3. Don't retry `voice_failed` even though it's a 503. Create a new voice from a better clip\n4. Use a key from the project that owns the voice, voices are per project\n5. Keep the clip under 2 minutes and 35 MB, longer fails with `voice_audio_too_long`\n6. Send the API key with the WebSocket connection, not in the config message, before 15 January 2027\n\n## Tutorials\n\n- [Prepare a project voice for Soniox TTS](https://www.anchorterminal.com/tutorials/soniox-voice-cloning/), for Soniox Voice Cloning\n\n## Questions\n\n### Which is better for AI agents, Gradium Voice Cloning and Voice Design or Soniox Voice Cloning?\n\nSoniox Voice Cloning scores 68.7 (B) on agent readiness against Gradium Voice Cloning and Voice Design's 60.8 (C), and leads in 6 of 7 scored categories. Gradium Voice Cloning and Voice Design leads on payments \u0026 pricing.\n\n### Do Gradium Voice Cloning and Voice Design and Soniox Voice Cloning need an API key?\n\nBoth need an API key.\n\n### Can an agent call Gradium Voice Cloning and Voice Design and Soniox Voice Cloning without installing anything?\n\nYes. Gradium Voice Cloning and Voice Design has a hosted endpoint at https://api.gradium.ai/api/voices/ and Soniox Voice Cloning at https://api.soniox.com/v1.\n\n\n## For agents\n\n- This comparison as JSON: https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-soniox-voice-cloning.json, and with the fewest tokens: https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-soniox-voice-cloning.min.md\n- Over MCP at https://www.anchorterminal.com/mcp (no key): `compare_tools {\"a\": \"gradium-voice-cloning\", \"b\": \"soniox-voice-cloning\"}`. From a terminal: `anchor compare gradium-voice-cloning soniox-voice-cloning`\n- Each listing in full: https://www.anchorterminal.com/api/v1/tools/gradium-voice-cloning.json and https://www.anchorterminal.com/api/v1/tools/soniox-voice-cloning.json\n\n## Other comparisons with Gradium Voice Cloning and Voice Design or Soniox Voice Cloning\n\n- [Cartesia Voice Cloning API + MCP vs Gradium Voice Cloning and Voice Design](https://www.anchorterminal.com/compare/cartesia-voice-cloning-vs-gradium-voice-cloning.md)\n- [Cartesia Voice Cloning API + MCP vs Soniox Voice Cloning](https://www.anchorterminal.com/compare/cartesia-voice-cloning-vs-soniox-voice-cloning.md)\n- [ElevenLabs Voice Cloning and Voice Design API vs Gradium Voice Cloning and Voice Design](https://www.anchorterminal.com/compare/elevenlabs-voice-cloning-vs-gradium-voice-cloning.md)\n- [ElevenLabs Voice Cloning and Voice Design API vs Soniox Voice Cloning](https://www.anchorterminal.com/compare/elevenlabs-voice-cloning-vs-soniox-voice-cloning.md)\n- [Fish Audio Voice Cloning API vs Gradium Voice Cloning and Voice Design](https://www.anchorterminal.com/compare/fish-audio-voice-cloning-vs-gradium-voice-cloning.md)\n- [Fish Audio Voice Cloning API vs Soniox Voice Cloning](https://www.anchorterminal.com/compare/fish-audio-voice-cloning-vs-soniox-voice-cloning.md)\n- [Gradium Voice Cloning and Voice Design vs Hume Octave Voice Design and Cloning + MCP](https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-hume-voice-cloning.md)\n- [Gradium Voice Cloning and Voice Design vs Inworld Instant Voice Cloning](https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-inworld-voice-cloning.md)\n- [Gradium Voice Cloning and Voice Design vs MiniMax Voice Clone and Voice Design](https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-minimax-voice-cloning.md)\n- [Gradium Voice Cloning and Voice Design vs Murf Voice Cloning API](https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-murf-voice-cloning.md)\n- [Gradium Voice Cloning and Voice Design vs PlayHT Voice Cloning API](https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-playht-voice-cloning.md)\n- [Gradium Voice Cloning and Voice Design vs Resemble AI Voice Cloning API](https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-resemble-ai-voice-cloning.md)\n- [Gradium Voice Cloning and Voice Design vs Smallest.ai Instant Voice Clone](https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-smallest-voice-clone.md)\n- [Gradium Voice Cloning and Voice Design vs Speechify API Voice Cloning](https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-speechify-voice-cloning.md)\n- [Gradium Voice Cloning and Voice Design vs Ultravox Voice Cloning](https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-ultravox-voice-cloning.md)\n- [Gradium Voice Cloning and Voice Design vs VoiceStudio](https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-voicestudio.md)\n- [Hume Octave Voice Design and Cloning + MCP vs Soniox Voice Cloning](https://www.anchorterminal.com/compare/hume-voice-cloning-vs-soniox-voice-cloning.md)\n- [Inworld Instant Voice Cloning vs Soniox Voice Cloning](https://www.anchorterminal.com/compare/inworld-voice-cloning-vs-soniox-voice-cloning.md)\n- [MiniMax Voice Clone and Voice Design vs Soniox Voice Cloning](https://www.anchorterminal.com/compare/minimax-voice-cloning-vs-soniox-voice-cloning.md)\n- [Murf Voice Cloning API vs Soniox Voice Cloning](https://www.anchorterminal.com/compare/murf-voice-cloning-vs-soniox-voice-cloning.md)\n- [PlayHT Voice Cloning API vs Soniox Voice Cloning](https://www.anchorterminal.com/compare/playht-voice-cloning-vs-soniox-voice-cloning.md)\n- [Resemble AI Voice Cloning API vs Soniox Voice Cloning](https://www.anchorterminal.com/compare/resemble-ai-voice-cloning-vs-soniox-voice-cloning.md)\n- [Smallest.ai Instant Voice Clone vs Soniox Voice Cloning](https://www.anchorterminal.com/compare/smallest-voice-clone-vs-soniox-voice-cloning.md)\n- [Soniox Voice Cloning vs Speechify API Voice Cloning](https://www.anchorterminal.com/compare/soniox-voice-cloning-vs-speechify-voice-cloning.md)\n- [Soniox Voice Cloning vs Ultravox Voice Cloning](https://www.anchorterminal.com/compare/soniox-voice-cloning-vs-ultravox-voice-cloning.md)\n- [Soniox Voice Cloning vs VoiceStudio](https://www.anchorterminal.com/compare/soniox-voice-cloning-vs-voicestudio.md)\n", "meta": { "attribution": "Anchor Terminal (https://www.anchorterminal.com)", "attributionRequired": true, "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)", "docs": "https://www.anchorterminal.com/docs/", "generatedAt": "2026-11-07", "license": "CC-BY-4.0", "method": "https://www.anchorterminal.com/benchmark/", "methodology": "0.7", "openapi": "https://www.anchorterminal.com/openapi.json", "preview": false, "run": "2026-10-01", "runLabel": "October 2026 research run", "terms": "https://www.anchorterminal.com/terms/#data" }, "page": { "breadcrumbs": [ { "name": "Home", "url": "https://www.anchorterminal.com/" }, { "name": "Compare", "url": "https://www.anchorterminal.com/compare/" }, { "name": "Gradium Voice Cloning and Voice Design vs Soniox Voice Cloning", "url": "" } ], "description": "Soniox Voice Cloning scores 68.7 (B) to Gradium Voice Cloning and Voice Design's 60.8 (C) for voice cloning. Prices, MCP, x402, uptime and agent notes side by side.", "facts": [ "Gradium Voice Cloning and Voice Design C 60.8", "Soniox Voice Cloning B 68.7", "scores" ], "h1": "Gradium Voice Cloning and Voice Design vs Soniox Voice Cloning", "image": "https://www.anchorterminal.com/assets/og/default.png", "path": "/compare/gradium-voice-cloning-vs-soniox-voice-cloning", "published": "2026-10-01", "section": "tools", "title": "Gradium Voice Cloning and Voice Design vs Soniox Voice Cloning (2026)", "toc": null, "updated": "2026-11-07", "url": "https://www.anchorterminal.com/compare/gradium-voice-cloning-vs-soniox-voice-cloning" }, "tokens": { "markdown": 5650, "slim": 3280 }, "version": 1 }