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

Azure AI Speech speech-to-text vs Deepgram Speech-to-Text (Nova-3, Flux)

Deepgram Speech-to-Text (Nova-3, Flux) and Azure AI Speech speech-to-text score within a point of each other on agent readiness, 76.9 (BB) and 76.3 (BB). Azure AI Speech speech-to-text leads on reliability, security & auth, maintenance & community and transparency & trust. Both do speech-to-text.

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

Which one, for what

Azure AI Speech speech-to-text BB

Best for Teams on Azure who need several modes (real time, synchronous files, cheap batch, custom models) under one resource, or strict default data handling.

Ahead on

  • Reliability, 90 against 80
  • Security & auth, 85 against 72
  • Maintenance & community, 80 against 73
  • Transparency & trust, 86 against 78

Watch for

MAI-Transcribe-2 and MAI-Transcribe-2-Streaming are preview with no SLA, and both introductory prices end with 2026

Deepgram Speech-to-Text (Nova-3, Flux) BB

Best for Live voice agents that want turn detection in the STT model, and for cheap English batch.

Ahead on

  • Schema & documentation, 95 against 80
  • Payments & pricing, 60 against 30

Also in its favour

  • Runs on your own machine
  • Free to start without a card

Watch for

Training on audio is the default and the opt-out is a per-request flag

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.

Payments & pricing Deepgram Speech-to-Text (Nova-3, Flux) +30

Azure AI Speech speech-to-text 30

No x402, MPP or L402 (0). Per-hour prices published without a login, though the page needs JavaScript and the Retail Prices API is the readable source (20). The F0 tier gives 5 real-time hours a month but an Azure subscription needs a card (0). A person signs up in a browser (0). MAI-Transcribe-2-Streaming's introductory $0.54 an hour appears only in Microsoft AI's launch post. No Azure meter for it was found in the Retail Prices API on 5 October, 17 October or 1 November, which doesn't move this line because the other per-hour prices are published. The East US meters for real time ($1), fast ($0.36) and batch ($0.18) an hour were unchanged on 1 November. 0 + 20 + 0 + 0 = 20. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 20 is 30.

Deepgram Speech-to-Text (Nova-3, Flux) 60

No x402, MPP or L402 (0). Per-minute prices for every model and add-on published without a login (20). $200 credit with no card (20). A person signs up in a browser to get the first key (0). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 40 is 60.

Schema & documentation Deepgram Speech-to-Text (Nova-3, Flux) +15

Azure AI Speech speech-to-text 80

The speech-to-text OpenAPI document for GA version 2025-10-15 is in Microsoft's public REST API specs, re-read on 1 November with 67 operations (25). No llms.txt found (0). Operation descriptions state purpose, and the speech-to-text overview says when to use real time, fast or batch (15). Batch request bodies are typed with enums and required fields, but fast transcription takes its options as a JSON string in the multipart definition field, documented but untyped on the wire (10). The REST reference carries examples and error responses per operation, per the 5 October check, not re-read on 1 November (15). Dated api-version values (current GA 2025-10-15, on the versions page updated 19 October) and monthly release notes (15). MAI-Transcribe-2-Streaming's WebSocket events are documented in prose and examples only, with no machine-readable contract, which leaves the graded REST surface unchanged. 25 + 0 + 15 + 10 + 15 + 15 = 80.

Deepgram Speech-to-Text (Nova-3, Flux) 95

OpenAPI file at developers.deepgram.com/openapi.json and an AsyncAPI file for the sockets (25). llms.txt (10). Reference pages state what each option does, and the docs say to use Flux for live agents and Nova-3 for files, though few options say when not to use them (15). Query parameters are typed with enums for model, language and encoding (15). Examples on every endpoint and a documented error format (15). Versioned /v1 and /v2 paths and a dated changelog (15).

Security & auth Azure AI Speech speech-to-text +13

Azure AI Speech speech-to-text 85

Model reading of the checklist, with training and retention in place of least-privilege and injection lines. Two regenerable resource keys for rotation, or Microsoft Entra ID tokens with Azure role-based access, which Microsoft recommends (30). Less 10 because the Realtime route for MAI-Transcribe-2-Streaming, added on 1 October 2026, documents an api-key query-string parameter on the WebSocket URI, using the same resource key, still so on the page updated 7 October and re-read on 1 November. The speech-to-text REST, fast, batch and Speech SDK docs take the key in a header only, so the deduction is applied once. SAS URLs for batch audio are scoped tokens for the customer's own storage, not the Speech credential, and aren't deducted (20). Microsoft doesn't use customer audio to train its models, per the 1 October check, not re-read since (20). Real-time and fast transcription audio isn't stored, and batch output stays until deleted or timeToLive expires, per the data privacy page updated 2026-08-26 and re-read on 1 November (15). Azure Monitor and the activity log record resource actions, though we didn't confirm per-request Speech logging (10). The MSRC disclosure policy, Microsoft's Azure bounty programme, SOC 2 and ISO 27001 reports, and public advisories. An MSRC search on 1 November found no CVE for the Azure Speech service (the Speech entries are Windows Speech components and Chromium). The microsoft.com security.txt passed its Expires date on 2026-09-23 and was still expired on 1 November, which we note without deducting, as on 5 and 17 October (20). 20 + 20 + 15 + 10 + 20 = 85.

Deepgram Speech-to-Text (Nova-3, Flux) 72

Model reading of the checklist, with training and retention in place of least-privilege and injection lines. Keys belong to a project, carry a role and can expire on a date or after a set duration. Browser JWTs live 30 seconds and travel in a header (30). Requests may be kept for model training under the Model Improvement Program unless each request sets mip_opt_out=true, which any account can do (10). Opted-out requests are kept only for processing, so zero retention is self-serve but not the default (10). Per-request logs and usage data in the console and API, no audit log of key actions found (10). SOC 2 Type 1 and Type 2 (certificates on request) and HIPAA with a BAA on the compliance page. deepgram.com now serves a valid security.txt (Contact security@deepgram.com, Expires 30 September 2027), not found on 30 September. No disclosure policy, bug bounty or public advisories found. The 1 October check gave this line 5 where deepgram-tts and deepgram-voice-agent gave 10 and 12 on the same compliance page, so it's read as deepgram-tts was on 1 November (12 of 20). 30 + 10 + 10 + 10 + 12 = 72.

Reliability Azure AI Speech speech-to-text +10

Azure AI Speech speech-to-text 90

Azure status page with post-incident reviews (20). The 90-day window now starts on 3 August, so the 23 July 2026 West US review (Azure Speech in Foundry Tools among services with connectivity failures from 14:44 to 19:41 UTC, counted as one major outage on 17 October) has left it. Two reviews remain on 1 November. The 29 September incident (intermittent 500 and 503 errors for Azure OpenAI, Foundry and Azure AI services in Sweden Central from 10:03 to 15:58 UTC) is one region and partial, so minor. The 30 September to 1 October gateway review (20:30 to 02:15 UTC) names networking services, not Speech. Nothing newer is listed. Minor incidents only (20). Limits published with numbers, 100 concurrent real-time requests and 600 fast or batch requests a minute by default, on the quotas page updated 23 October and re-read on 1 November (15). 429 documented as usually meaning autoscaling in progress, with retry logic advised and a 1-2-4-4 minute pattern for ramping load. Fast transcription is synchronous, so a retry doesn't duplicate a job (15). Covered by Microsoft's online services SLA, per the 1 October check (10). Fast transcription, batch and real time are GA. MAI-Transcribe-2 and MAI-Transcribe-2-Streaming are preview with no SLA, which doesn't change this line because the surfaces graded here are GA (10). 20 + 20 + 15 + 15 + 10 + 10 = 90.

Deepgram Speech-to-Text (Nova-3, Flux) 80

Statuspage at status.deepgram.com with component history and an RSS feed (20). Nine incidents in the 90 days from 3 August touch speech-to-text, every one partial, read as this run reads partial incidents (one model, one region, one datacenter or some requests) and as deepgram-tts is read. Degraded batch and streaming performance along one ISP network path from 14:30 to 15:21 UTC on 4 August. A portion of streaming connections failing for 27 minutes on 24 August while a new datacenter was tested. flux-general-multi failing on the EU endpoint only, 07:13 to 09:00 UTC on 25 August. Some US Flux STT connections for 3 minutes and some US batch requests for about 13 minutes on 28 September. Two 1 to 2 minute bursts on 30 September. Some batch requests slow or timing out, 19:27 to 21:53 UTC on 22 October. 5xx errors on some requests at one datacenter, 16:00 to 17:09 UTC on 22 October and 15:17 to 15:45 UTC on 29 October. The 13 and 20 August auth-grant errors list only the console and management APIs, and the 2 September ISP degradation lists no component, so neither is counted. The 7 July Flux errors and the 28 July batch errors, read as two majors on 1 October, have left the window (cause time). Minor only (20). Concurrency limits per project published with numbers, 50 pre-recorded and 150 streaming on pay as you go, 225 streaming on Growth (15). Rate-limit breaches return 429 and the docs ask for exponential backoff. Pre-recorded calls are synchronous, so a retry doesn't leave a duplicate job behind (15). No SLA found for self-serve plans (0). Nova-3 and Flux are GA (10). 20 + 20 + 15 + 15 + 0 + 10 = 80.

Transparency & trust Azure AI Speech speech-to-text +8

Azure AI Speech speech-to-text 86

Closed service under Microsoft's product terms, with an MIT samples repository (15). The data privacy page, the privacy statement and the product terms agree on no storage for real time, retention until deletion for batch and no training (25). The data privacy page, re-read on 1 November and still dated 26 August, doesn't name the MAI streaming model. Dated retirements, REST API v3.0 and the v3.2 previews on 2026-03-31 and MAI-Transcribe-1 deprecated on 2026-08-20, under Microsoft's published lifecycle policy. The versions page now says 2024-05-15-preview will be retired on a date to be announced (20). Regions chosen per resource and a public sub-processor list (20). 15 + 25 + 20 + 20 = 80.

Deepgram Speech-to-Text (Nova-3, Flux) 78

Closed service with clear terms (15). The privacy policy was last updated on 26 October 2021, says data is stored on US servers and doesn't mention the Model Improvement Program, while the docs describe training by default and EU, Australian and Indian endpoints, so the statements disagree (15). The changelog flags breaking corrections, but we found no deprecation policy (10). Sub-processor list at deepgram.com/privacy/subprocessors and four regional endpoints (20). Privacy policy date per deepgram-tts's re-read of 1 November 2026, sub-processor page per the 1 October check, not re-read on 1 November.

Maintenance & community Azure AI Speech speech-to-text +7

Azure AI Speech speech-to-text 80

Speech SDK 1.53.0 is the latest on npm on 1 November (the JavaScript SDK's v1.53.0 tag is dated 21 October, and the samples repository moved to 1.53.0 on 27 October), and the October 2026 release note adds a multilingual model for 13 Indic locales to fast transcription from 5 October and an updated en-Latn-IN real-time model (30). Release notes for August, September and October 2026, and SDK 1.52.0 (September) and 1.53.0 (October) (20). Public release notes, a GitHub samples repository and Microsoft Q&A (10). Speech SDK current in eight languages (15). The SDK itself is a closed binary, so we can't see its CI (5). 30 + 20 + 10 + 15 + 5 = 80.

Deepgram Speech-to-Text (Nova-3, Flux) 73

Read as a model, as deepgram-tts is read on the same changelog and SDK. Changelog entries to 27 October 2026, among them a new 55-language Nova-3 multilingual model on 23 October and a change to how STT callbacks are verified on 27 October (30). 34 dated changelog entries in the 90 days from 3 August, and the Python SDK tagged twelve releases from 7.7.0 on 12 August to 7.16.0 on 28 October (10 of 10). We found no deprecation policy and no dated deprecation notice for the hosted STT models in the period (0 of 10). Public changelog and community support channels (10 of 15). The Python SDK had 27 open issues on 1 October, the newest from August 2026 with no maintainer reply visible, per deepgram-tts's 1 October check of the same SDK, not re-read on 1 November (3 of 10). Official SDKs current, the Python SDK at 7.16.0 on 28 October (15). SDK CI not checked in this run (5 of 10). 30 + 10 + 0 + 10 + 3 + 15 + 5 = 73.

Score by category

CategoryWeightAzure AI Speech speech-to-textDeepgram Speech-to-Text (Nova-3, Flux)Edge
Reliability16%9080Azure AI Speech speech-to-text +10
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%8095Deepgram Speech-to-Text (Nova-3, Flux) +15
Agent ergonomics13%7575even
Security & auth14%8572Azure AI Speech speech-to-text +13
Payments & pricing10%3060Deepgram Speech-to-Text (Nova-3, Flux) +30
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8073Azure AI Speech speech-to-text +7
Transparency & trust7%8678Azure AI Speech speech-to-text +8
Negative events≤1500
TotalBB 76.3/100BB 76.9/100

Facts side by side

FactAzure AI Speech speech-to-textDeepgram Speech-to-Text (Nova-3, Flux)
KindModel APIModel API
VendorMicrosoftDeepgram
Hosted endpointhttps://eastus.api.cognitive.microsoft.com/speechtotexthttps://api.deepgram.com/v1
TransportsHTTPHTTP, Streamable HTTP, stdio, SSE (legacy)
AuthOAuth or keyAPI key
PricingFreemiumPay per use
x402nono
LicenceMIT (samples), SDK under Microsoft's own licenceMIT (SDKs)
Read-only variant documentednono
llms.txtnoyes
Last release2026-10-212026-10-28
Terms last updatedcouldn't be read2026-08-06
Privacy policy last updated2026-09-012021-10-26
Customer content may train modelsyesyes, with an opt-out
Terms restrict automated accesscouldn't be readnot found in the text
Terms restrict benchmarkingcouldn't be readyes
Terms or service can change without noticecouldn't be readyes
Arbitration or class-action waivercouldn't be readyes
Popularity3.5k stars, 476k npm/wk, 1M PyPI/wk468 stars, 1.1M npm/wk, 805k PyPI/wk
Agent reviews3.3/5 (8)3.5/5 (2)

The panel on both

The 2 reviewers who reviewed both split evenly.

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

TaskAzure AI Speech speech-to-textDeepgram Speech-to-Text (Nova-3, Flux)
Transcribe a one-hour meeting60 minutes of recorded audio$0.10 to $1.20$0.26 to $0.55

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

Azure AI Speech speech-to-text

Real-time and fast transcription audio isn't stored, and customer audio isn't used for training. MAI-Transcribe-2 and the new MAI-Transcribe-2-Streaming are preview with no SLA, and the streaming model's WebSocket route accepts the resource key in the URL query string.

Deepgram Speech-to-Text (Nova-3, Flux)

Flux streams with model-level end-of-turn detection, so a voice agent needs no separate VAD. Training on audio is the default and the opt-out is a per-request flag.

Before you call either

Azure AI Speech speech-to-text

  1. Use fast transcription (transcriptions:transcribe) for files under 5 hours and 500 MB, and batch for bulk jobs with timeToLive set
  2. Pin api-version=2025-10-15. v3.0 and the v3.2 previews are retired
  3. On a 429, back off 1, 2, 4 then 4 minutes. It usually means autoscaling, not a quota
  4. For MAI-Transcribe-2-Streaming, use Speech SDK 1.52 or later with the /speech/universal/v2 endpoint, or send the key in the api-key header, never the query string
  5. Don't budget on MAI-Transcribe-2 at $0.10 or MAI-Transcribe-2-Streaming at $0.54 an hour after 2026-12-31

Deepgram Speech-to-Text (Nova-3, Flux)

  1. Add mip_opt_out=true to every request that carries customer audio
  2. Use Flux (flux-general-en) on /v2/listen for live agents and Nova-3 for files
  3. Back off exponentially on 429. The concurrency limit is per project
  4. Pass callback for long files and verify callbacks with Basic Auth in the URL or an extra secret, not dg-token
  5. Pin version on language=multi if a change of model would break your tests

Questions

Which is better for AI agents, Azure AI Speech speech-to-text or Deepgram Speech-to-Text (Nova-3, Flux)?

Deepgram Speech-to-Text (Nova-3, Flux) and Azure AI Speech speech-to-text score within a point of each other on agent readiness, 76.9 (BB) and 76.3 (BB). Azure AI Speech speech-to-text leads on reliability, security & auth, maintenance & community and transparency & trust.

Do Azure AI Speech speech-to-text and Deepgram Speech-to-Text (Nova-3, Flux) need an API key?

Azure AI Speech speech-to-text takes an API key or an OAuth sign-in. Deepgram Speech-to-Text (Nova-3, Flux) needs an API key.

Can an agent call Azure AI Speech speech-to-text and Deepgram Speech-to-Text (Nova-3, Flux) without installing anything?

Yes. Azure AI Speech speech-to-text has a hosted endpoint at https://eastus.api.cognitive.microsoft.com/speechtotext and Deepgram Speech-to-Text (Nova-3, Flux) at https://api.deepgram.com/v1.

Other comparisons with Azure AI Speech speech-to-text or Deepgram Speech-to-Text (Nova-3, Flux)

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.