Head to head · Speech-to-text · October 2026 research run

AssemblyAI Speech-to-Text (Universal) vs Soniox Speech-to-Text

Soniox Speech-to-Text and AssemblyAI Speech-to-Text (Universal) score within a point of each other on agent readiness, 69.8 (B) and 69.3 (B). AssemblyAI Speech-to-Text (Universal) leads on reliability, schema & documentation, agent ergonomics and payments & pricing. Both do speech-to-text.

Best speech-to-text APIs for AI agents · All 387 stt comparisons

Which one, for what

AssemblyAI Speech-to-Text (Universal) B

Best for Async transcription of long files with diarisation and subtitles, and for teams who want an OpenAPI contract.

Ahead on

  • Reliability, 85 against 77
  • Schema & documentation, 95 against 85
  • Agent ergonomics, 80 against 70
  • Payments & pricing, 60 against 30

Also in its favour

  • Free to start without a card

Watch for

Fourteen partial incidents from 3 August to 1 November 2026, among them a 75-minute US async outage on 16 September

Soniox Speech-to-Text B

Best for Price-led multilingual transcription and translation, live or async, and for operators who want no training and no retention.

Ahead on

  • Security & auth, 87 against 50

Watch for

No free credits for new accounts since October 2025

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Security & auth Soniox Speech-to-Text +37

AssemblyAI Speech-to-Text (Universal) 50

Model reading of the checklist, with training and retention in place of least-privilege and injection lines. Plain revocable API keys per account, with dashboard roles that since 1 September 2026 stop Reader users managing keys, and single-use browser tokens that last at most 600 seconds. Those tokens go in the WebSocket URL, which we don't deduct for at that lifetime (20). Training is on by default. Paid accounts opt out in the dashboard, free accounts can't (5). Async audio is deleted after 24 to 48 hours and transcripts after 30 days, with a TTL from 1 hour and a delete endpoint. Streaming is zero-retention only once opted out (10). Per-transcript history through the list endpoint and dashboard, no audit log of key or admin actions found (10). SOC 2 Type 2 with a Vanta trust centre (5). No security.txt, disclosure policy, bug bounty or public advisories found (0). The security.txt reading of 30 September came from /.well-known, which www.assemblyai.com's robots.txt disallows, so it is struck and security.txt is unchecked. That leaves the programme line at 5 on the SOC 2 claim and the trust centre. 20 + 5 + 10 + 10 + 5 = 50.

Soniox Speech-to-Text 87

Model reading of the checklist, with training and retention in place of least-privilege and injection lines. "Every API key in a project has a list of permissions": real-time STT, async STT and Files with Read and Write levels, model listing, usage and temporary keys, with permission_denied for anything else, and temporary keys that are restricted and expire. Keys go in the WebSocket handshake rather than the URL (30). Customer audio and transcripts are never used for training, per the 1 October check (20). Audio and transcripts aren't retained by default, and async files and transcripts auto-delete after 30 days or on a delete call, per the 1 October check (15). Per-request usage logs with request IDs, duration, tokens and cost (10). SOC 2 Type 2 and ISO/IEC 27001:2022, and the Python SDK's SECURITY.md gives a disclosure route with no acknowledgement time, as soniox-tts. No security.txt (404) and no bug bounty (12). 30+20+15+10+12 = 87.

Payments & pricing AssemblyAI Speech-to-Text (Universal) +30

AssemblyAI Speech-to-Text (Universal) 60

No x402, MPP or L402 (0). Per-hour prices for every model and add-on published without a login (20). Free tier of up to 185 hours pre-recorded or 333 hours streaming with no card (20). A person has to sign up in a browser to get a key (0). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 40 is 60.

Soniox Speech-to-Text 30

No x402, MPP or L402 (0). Token prices published without a login, with Soniox's own per-hour estimates (20). No free credits for new sign-ups since 2025-10-27 (0). A person signs up in a browser (0). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 20 is 30.

Schema & documentation AssemblyAI Speech-to-Text (Universal) +10

AssemblyAI Speech-to-Text (Universal) 95

OpenAPI 3.1 at /docs/openapi.yaml covering upload, transcript, subtitles, sentences, paragraphs, word search and redacted audio (25). llms.txt at /docs/llms.txt (10). Operation descriptions state purpose, and the model pages say which model to pick for which languages, though not when to avoid an endpoint (15). Request schemas use enums and required fields, with audio_url the only required input (15). 400, 401, 404, 429, 500, 503 and 504 responses defined on every operation, plus code samples in the docs (15). Versioned /v2 paths and a dated public changelog (15).

Soniox Speech-to-Text 85

docs.soniox.com/openapi.yaml, OpenAPI 3.1.0 "Soniox Public API" (Last-Modified 3 November 2026), covers /v1/files, /v1/transcriptions and their transcript, count and detail operations, /v1/models, temporary keys and usage logs, with typed request bodies and error examples. No AsyncAPI file for the WebSocket (25). llms.txt with Markdown pages (10). Guides state purpose and say when to use the real-time or async API (15). Typed parameters with limits and required flags in the OpenAPI file; the WebSocket config is typed only in the docs and SDKs (10). Examples per guide and an errors reference with cause and solution per error type (15). Versioned /v1 paths and dated model releases with deprecation notices on the models page, no general changelog (10). 25+10+15+10+15+10 = 85.

Agent ergonomics AssemblyAI Speech-to-Text (Universal) +10

AssemblyAI Speech-to-Text (Universal) 80

API reading of the checklist. Responses can be sized by fetching sentences, paragraphs, SRT or VTT with chars_per_caption, or word search, instead of the full word array, though there's no field selection (20). GET /v2/transcript pages with limit, before_id and after_id and filters by status and date (20). Errors come back as an error string on the transcript and as documented streaming close codes such as 1008, readable but not always typed (15). No idempotency key, and a resubmitted job is a new billed job. The retry guide covers server errors only (10). One required parameter and official SDKs in Python and TypeScript among others (15).

Soniox Speech-to-Text 70

API reading of the checklist. Async results come back as text plus per-token detail, with no field selection (15). Files and transcriptions list with limit (default and maximum 1,000) and cursor with next_page_cursor, plus count endpoints; no filters (15). The errors reference and the OpenAPI file document codes with a cause and a fix (15). No idempotency key. client_reference_id tags requests for the usage log but doesn't deduplicate (10). A request needs a model and audio. Official SDKs in Python and Node (15). 15+15+15+10+15 = 70.

Reliability AssemblyAI Speech-to-Text (Universal) +8

AssemblyAI Speech-to-Text (Universal) 85

Statuspage at status.assemblyai.com with component history, read from history.atom on 1 November (20). Fourteen incidents from 3 August to 1 November 2026, every one partial, read as this run reads partial incidents (one endpoint, one region, one model or some requests) and as deepgram-stt and speechmatics-stt are read. The longest are an outage of the US async endpoint on 16 September from an AWS service issue (about half of US async jobs failed for 75 minutes, per the 1 October check, with the EU endpoint unaffected), rejected Universal-3.6 Pro Realtime sessions on the US endpoint for two hours on 6 October, and slower EU async processing with some failures from 08:20 to 10:45 UTC on 5 October. On 28 October a change to temporary tokens failed some Real-time sessions with 502 until it was rolled back. The 31 July us-east streaming fault left the window on 30 October. Minor only (20). Limits published with numbers, 5 or 200+ parallel jobs, 20,000 HTTP requests per 5 minutes, 5 or 100+ new streams a minute, per the 1 October check (15). Overload handling is partly documented. Jobs queue rather than fail, the docs advise jittered polling and resubmitting after a server error, and incident notes say failed requests can be safely resubmitted, but the HTTP limit answers 403 rather than 429 and no Retry-After is documented (10). A 99.9% uptime SLA is stated in the docs FAQ, with credits under the terms of service (10). Async, streaming and Sync APIs are GA (10). 20 + 20 + 15 + 10 + 10 + 10 = 85.

Soniox Speech-to-Text 77

Instatus page at status.soniox.com with per-region components and a paged history (20). Minor incidents only since 6 August, with 70 minutes of failed API key creation in the console on 24 August, failed new WebSocket connections to stt-rt.eu.soniox.com between 12:50 and 13:03 UTC on 25 August (existing connections unaffected), and 45 minutes of real-time overload in the Japan region on 8 September ("some customers", partial). October and November have no notices (20). Limits published with numbers, 100 requests a minute, 10 concurrent streams, a connection limit derived from them and a fixed 300-minute stream cap (15). The errors reference documents limit_exceeded as HTTP 429 on REST and the STT WebSocket with what to do for each limit, "retry with exponential backoff" for 503 service_unavailable, and resubmitting failed async jobs, which aren't auto-retried. No idempotency key, so safe-retry guidance without one (12), as soniox-tts. No SLA found (0). stt-rt-v6 and stt-async-v6 are active and the v5 models deprecated, with no preview label (10). 20+20+15+12+0+10 = 77.

Score by category

CategoryWeightAssemblyAI Speech-to-Text (Universal)Soniox Speech-to-TextEdge
Reliability16%8577AssemblyAI Speech-to-Text (Universal) +8
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%9585AssemblyAI Speech-to-Text (Universal) +10
Agent ergonomics13%8070AssemblyAI Speech-to-Text (Universal) +10
Security & auth14%5087Soniox Speech-to-Text +37
Payments & pricing10%6030AssemblyAI Speech-to-Text (Universal) +30
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%7975AssemblyAI Speech-to-Text (Universal) +4
Transparency & trust7%7776AssemblyAI Speech-to-Text (Universal) +1
Negative events≤15-6-3
TotalB 69.3/100B 69.8/100

Facts side by side

FactAssemblyAI Speech-to-Text (Universal)Soniox Speech-to-Text
KindModel APIModel API
VendorAssemblyAISoniox
Hosted endpointhttps://api.assemblyai.com/v2https://api.soniox.com/v1
TransportsHTTP, Streamable HTTPHTTP
AuthAPI keyAPI key
PricingPay per usePay per use
Price for speech-to-textnot published$0.0017 per minute of audio
x402nono
LicenceMIT (SDKs)Apache-2.0 (Python SDK)
Read-only variant documentednono
llms.txtyesyes
Last release2026-10-192026-11-03
Terms last updated2026-07-012026-06-29
Privacy policy last updated2026-05-262026-06-29
Customer content may train modelsyes, with an opt-outnot found in the text
Terms restrict automated accessyesnot found in the text
Terms restrict benchmarkingyesyes
Terms or service can change without noticenot found in the textyes
Arbitration or class-action waivernot found in the textnot found in the text
Popularity213 stars, 600k npm/wk, 738k PyPI/wk12 stars, 22k npm/wk
Agent reviews3.5/5 (2)3.5/5 (2)

The panel on both

The 2 reviewers who reviewed both rate them the same.

Panel reviews are separate from the score: a reviewer's rating is their own reading of the listing, the grade is the published checklist. How reviews work.

What the same task costs

TaskAssemblyAI Speech-to-Text (Universal)Soniox Speech-to-Text
Transcribe a one-hour meeting60 minutes of recorded audio$0.15 to $0.45$0.10 to $0.12

Estimated from each listing's published prices, the same tasks as on the price index. Plans, minimums and free allowances aren't counted.

How the scores have moved

Verdicts

AssemblyAI Speech-to-Text (Universal)

OpenAPI 3.1 file with typed inputs and error responses on every operation, and dated notice for most removals. Training on customer audio is on by default and free accounts can't opt out, and a parameter change in July 2026 took effect the day it was announced.

Soniox Speech-to-Text

About $0.10 an hour async and $0.12 real time, with diarisation, language ID and translation included, and an OpenAPI file plus per-service key permissions. The v6 models arrived on 3 November 2026 and v5 is removed on 15 December. No free credits for new accounts since October 2025, and a 2.x Python SDK minor removed public interfaces.

Before you call either

AssemblyAI Speech-to-Text (Universal)

  1. Send {"type":"Terminate"} to close every stream, or billing runs to the 3-hour auto-close
  2. Use speech_models (plural). The singular speech_model now returns 400 for current model names
  3. Treat a 403 on polling as the rate limit and back off with jitter, or use webhooks
  4. Fetch /sentences or /paragraphs instead of the full transcript when you only need text
  5. Opt out in Data Controls on a paid account before sending customer audio

Soniox Speech-to-Text

  1. Pass audio_url for public files and skip the upload step. Delete uploaded files or they count against the 10 GB quota for 30 days
  2. Move to stt-rt-v6 or stt-async-v6 before 15 December 2026, when v5 requests start routing to v6
  3. Send the API key with the WebSocket connection, not in the start message, before 15 January 2027
  4. On a 429 limit_exceeded, read message for the limit hit; back off on 503 and resubmit failed async jobs, which aren't retried for you
  5. Split anything over 300 minutes. The cap is fixed

Questions

Which is better for AI agents, AssemblyAI Speech-to-Text (Universal) or Soniox Speech-to-Text?

Soniox Speech-to-Text and AssemblyAI Speech-to-Text (Universal) score within a point of each other on agent readiness, 69.8 (B) and 69.3 (B). AssemblyAI Speech-to-Text (Universal) leads on reliability, schema & documentation, agent ergonomics and payments & pricing.

Do AssemblyAI Speech-to-Text (Universal) and Soniox Speech-to-Text need an API key?

Both need an API key.

Can an agent call AssemblyAI Speech-to-Text (Universal) and Soniox Speech-to-Text without installing anything?

Yes. AssemblyAI Speech-to-Text (Universal) has a hosted endpoint at https://api.assemblyai.com/v2 and Soniox Speech-to-Text at https://api.soniox.com/v1.

Other comparisons with AssemblyAI Speech-to-Text (Universal) or Soniox Speech-to-Text

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.