Soniox Text-to-Speech by Soniox

# Fix list: Soniox Text-to-Speech

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/soniox-tts, the October 2026 research run, assessed 1 October 2026. Grade B, 63.9 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on Soniox Text-to-Speech: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Payments & pricing, 20 out of 100, up to 10 more on the total

Why it scored 20: No x402, MPP or L402 (0). Token prices published without a login, $4.00 per 1M input tokens and $21.50 per 1M output tokens, which Soniox puts at about $0.70 an hour (20). No free credits for new accounts found (0). A person signs up and funds the account in a browser (0).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 2. Schema & documentation, 60 out of 100, up to 6.5 more on the total

Why it scored 60: No OpenAPI or AsyncAPI file found (0). llms.txt, llms-full.txt and `.mdx` pages (10). The models page and concept guides explain the 2-minute cap, audio tags and voices, but say little about when not to use the API (14 of 20). REST and WebSocket parameters are typed in the reference, with formats listed, but there's no machine-readable schema to check (11 of 15). One error format across surfaces with a stable `error_type`, a `request_id` and a `more_info` link, more than 25 types including TTS ones such as `max_audio_duration_reached` (15). Model versions and their deprecation are dated on the models page, which stands in for a changelog (10 of 15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 3. Agent ergonomics, 68 out of 100, up to 5.2 more on the total

Why it scored 68: API reading of the checklist. PCM in four encodings, WAV, MP3, Opus, AAC and FLAC, with timestamps on the WebSocket (22 of 25). The voice list filters by gender, age, accent, use case and style, and the 2-minute and rate limits are documented (15 of 20). Machine-readable `error_type` values an agent can branch on (20). No retry guidance and nothing found on billing for failed or truncated calls (0 of 20). Model, language, voice, format and text are all set per request. SDKs for Python, Node, the web and React (11 of 15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 4. Security & auth, 75 out of 100, up to 4.4 more on the total

Why it scored 75: Model reading of the checklist, with training and retention in place of least-privilege and injection lines. Project API keys plus temporary keys for browser clients that expire (25 of 30). Soniox says audio and text are never used to improve its models (20). Nothing is stored unless you ask, and logs carry no content (15). Usage in the console, no per-request log found (5 of 15). SOC 2 Type 2, ISO/IEC 27001:2022 and HIPAA listed on the security page. No security.txt, disclosure policy or bug bounty found (10 of 20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 5. Maintenance & community, 50 out of 100, up to 4.4 more on the total

Why it scored 50: Read as a model. `tts-rt-v2` went GA on 2026-08-11 alongside Python SDK 2.9.0, 51 days ago (20). Two dated model events and one SDK release in the last 90 days (0 of 10). `tts-rt-v1` was deprecated on 2026-08-11 and removed on 2026-08-31, 20 days later, with traffic routed to v2 (0 of 10). The models page and a support channel, no SDK issue tracker checked (8 of 25). SDKs for Python, Node, the web and React (15). The Python SDK supports Python 3.10 to 3.14 and shipped six releases between 14 May and 11 August. CI not checked (7 of 10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## 6. Reliability, 83 out of 100, up to 3.4 more on the total

Why it scored 83: Instatus page at status.soniox.com with separate TTS REST and TTS Real-time components in the US, EU, Japan and India, and a dated history (20). The page shows 100 per cent uptime for every region over 90 days and no TTS incident in the history we could read, which starts in August. The three incidents on record hit real-time STT in Japan (45 minutes on 8 September), EU WebSocket connections for STT (9 minutes on 25 August) and key creation in the console (70 minutes on 24 August) (30). Limits published, 100 requests a minute and 3 concurrent requests or streams, raisable in the console (15). A 429 returns `limit_exceeded` with advice to slow down, but no backoff pattern or Retry-After (8). No SLA found (0). `tts-rt-v2` is GA in all four regions (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 7. Transparency & trust, 74 out of 100, up to 2.3 more on the total

Made of editorial 62, provenance 86.

Why it scored 74: Closed service under terms updated 2026-06-29, Python SDK under Apache-2.0 (15). The security page, the terms and the TTS docs agree on no storage and no training, with async data deleted after 30 days (25 of 30). Model removals are dated on the models page, with no written policy and 20 days' notice for v1 (10 of 20). Regional hosts in the US, EU, Japan and India and a data residency page, no sub-processor list found (12 of 20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

Provenance checks not met in full (half of this category, computed from checked facts):

- Domain age: soniox.com, registered 2020-03-23 (6 years) (11 of 15)
- security.txt: not found (0 of 10)

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- Whether new accounts get any free credit. The listing says none and the pricing page doesn't say.
- SDK issue responsiveness and CI, which we didn't check.
- Whether truncated or failed requests are billed.

## Weaknesses

- Audio stops at 2 minutes per request or stream and the cap can't be raised
- 3 concurrent requests and 100 requests a minute by default
- `tts-rt-v1` was removed 20 days after its deprecation notice
- No free credits for new accounts
- No OpenAPI file and no retry guidance

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- Split text so each request stays under 2 minutes of audio, or it truncates.
- Branch on `error_type`, not the message, and back off on `limit_exceeded`.
- Use a temporary API key for browser clients.
- Use bracketed audio tags such as `[whispering]` instead of SSML.
- Pick the regional host (EU, Japan, India) that matches your data residency.

## What the review panel asked for

- State billing for truncated requests
- add Retry-After to 429
- say whether truncated calls are billed

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.