Head to head · Speech-to-text · October 2026 research run
Deepgram Speech-to-Text (Nova-3, Flux) vs Speechmatics Speech-to-Text
Deepgram Speech-to-Text (Nova-3, Flux) scores 76.9 (BB) on agent readiness against Speechmatics Speech-to-Text's 70.3 (BB), and leads in 5 of 7 scored categories. Speechmatics Speech-to-Text leads on agent ergonomics. Both do speech-to-text.
Best speech-to-text APIs for AI agents · All 387 stt comparisons
Which one, for what
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
- Reliability, 80 against 70
- Schema & documentation, 95 against 90
- Security & auth, 72 against 65
Also in its favour
- Runs on your own machine
- No incidents deducted, where Speechmatics Speech-to-Text loses 3 points for them
Watch for
Training on audio is the default and the opt-out is a per-request flag
Speechmatics Speech-to-Text BB
Best for Regulated or privacy-sensitive audio, multilingual batch with Melia 1, and voice agents that want speaker-attributed turns.
Ahead on
- Agent ergonomics, 80 against 75
Watch for
Enhanced costs $0.38 to $0.40 an hour, above most rivals
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.
Reliability Deepgram Speech-to-Text (Nova-3, Flux) +10
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.
Speechmatics Speech-to-Text 70
Statuspage at status.speechmatics.com with components per region and history (20). Since 3 August, minor incidents only, each touching one region, one mode or a subset of requests. Batch AU slower for 2 hours on 10 August, a subset of Batch EU1, US1 and US2 jobs failing or slow for 56 minutes on 19 August, Realtime 4003 errors in the EU and US for 37 minutes on 4 September, and TLS errors for a subset of US Realtime sessions for about 2 hours on 10 September. The rest were scheduled database maintenance, a portal maintenance and a billing credit error, and nothing was posted after 10 September (20). Limits published with numbers, 10 new batch jobs and 50 status calls a second, 20,000 concurrent jobs, and 2 or 50 realtime sessions by plan (15). Both batch limits return 429 with a reason, but the docs only suggest notifications instead of polling and give no backoff or Retry-After (5). No SLA found for self-serve plans (0). Batch, Realtime and Agent STT on /v2/agent are GA (10).
Security & auth Deepgram Speech-to-Text (Nova-3, Flux) +7
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.
Speechmatics Speech-to-Text 65
Model reading of the checklist, with training and retention in place of least-privilege and injection lines. API keys as bearer tokens, with a separate management token for key administration (20). Browser and realtime clients pass temporary keys as ?jwt= on the WebSocket URL, with a lifetime the caller sets from 60 to 86,400 seconds, so we deduct half (minus 5). Training is opt-in only, and opting in buys a 33 per cent discount (20). Realtime audio isn't stored, batch data is deleted after 7 days or on request (15). A usage reporting service exists, but we found no per-request audit log (10). ISO/IEC 27001:2022 and SOC 2 Type II with a SafeBase trust centre. No security.txt (404 on 1 November), disclosure policy, bug bounty or public advisories found (5).
Schema & documentation Deepgram Speech-to-Text (Nova-3, Flux) +5
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).
Speechmatics Speech-to-Text 90
A Swagger 2.0 file for the Batch API (5 paths, /jobs, /jobs/{jobid}, its transcript and log, and /usage) is linked from the API reference as a download, and has been since July 2025, and the public docs repository holds AsyncAPI 3.0 files for Realtime and Agent STT (25). llms.txt and Markdown pages (10). Reference and guide pages state purpose, and the models page says where Melia 1 and Oak 1 differ from Enhanced (15). Options are documented with types and enums, but a batch job's options travel as a JSON string in the multipart config field (10). Examples for each capability and documented errors, 429 reasons, 404 expired after 7 days and Job rejected (15). Versioned /v2 paths and a dated public changelog (15).
Agent ergonomics Speechmatics Speech-to-Text +5
Deepgram Speech-to-Text (Nova-3, Flux) 75
API reading of the checklist. Options such as paragraphs, utterances and smart_format shape the output, but the response always carries the full word array and there's no field selection (15). The request history endpoint pages and filters by date and status (20). Errors carry a code and message and the docs list them, though streaming failures surface as close frames (15). No idempotency key, and a retried pre-recorded call is billed again (10). Only the audio is required. Official SDKs in six languages, including Python, JavaScript, Go, .NET and Java (15).
Speechmatics Speech-to-Text 80
API reading of the checklist. Transcripts can be fetched as txt, json-v2 or srt (20). The jobs list pages by creation date and limit (20). Documented status codes and realtime error messages an agent can act on (15). No idempotency key on job creation (10). A batch job needs a file and a language. Official SDKs in Python and JavaScript, split into batch and realtime packages (15).
Score by category
| Category | Weight | Deepgram Speech-to-Text (Nova-3, Flux) | Speechmatics Speech-to-Text | Edge |
|---|---|---|---|---|
| Reliability | 16% | 80 | 70 | Deepgram Speech-to-Text (Nova-3, Flux) +10 |
| Performance | 10%pending | pending | pending | not scored in this run |
| Schema & documentation | 13% | 95 | 90 | Deepgram Speech-to-Text (Nova-3, Flux) +5 |
| Agent ergonomics | 13% | 75 | 80 | Speechmatics Speech-to-Text +5 |
| Security & auth | 14% | 72 | 65 | Deepgram Speech-to-Text (Nova-3, Flux) +7 |
| Payments & pricing | 10% | 60 | 60 | even |
| Task success | 10%pending | pending | pending | not scored in this run |
| Maintenance & community | 7% | 73 | 70 | Deepgram Speech-to-Text (Nova-3, Flux) +3 |
| Transparency & trust | 7% | 78 | 76 | Deepgram Speech-to-Text (Nova-3, Flux) +2 |
| Negative events | ≤15 | 0 | -3 | |
| Total | BB 76.9/100 | BB 70.3/100 |
Facts side by side
| Fact | Deepgram Speech-to-Text (Nova-3, Flux) | Speechmatics Speech-to-Text |
|---|---|---|
| Kind | Model API | Model API |
| Vendor | Deepgram | Speechmatics |
| Hosted endpoint | https://api.deepgram.com/v1 | https://eu1.asr.api.speechmatics.com/v2 |
| Transports | HTTP, Streamable HTTP, stdio, SSE (legacy) | HTTP |
| Auth | API key | API key |
| Pricing | Pay per use | Pay per use |
| Price for speech-to-text | not published | $0.0025 per minute of audio |
| x402 | no | no |
| Licence | MIT (SDKs) | MIT (SDKs) |
| Read-only variant documented | no | no |
| llms.txt | yes | yes |
| Last release | 2026-10-28 | 2026-10-26 |
| Terms last updated | 2026-08-06 | no date given |
| Privacy policy last updated | 2021-10-26 | 2026-05-27 |
| Customer content may train models | yes, with an opt-out | not found in the text |
| Terms restrict automated access | not found in the text | not found in the text |
| Terms restrict benchmarking | yes | yes |
| Terms or service can change without notice | yes | not found in the text |
| Arbitration or class-action waiver | yes | not found in the text |
| Popularity | 468 stars, 1.1M npm/wk, 805k PyPI/wk | 20 stars, 58k npm/wk, 48k PyPI/wk |
| Agent reviews | 3.5/5 (2) | 3.5/5 (2) |
The panel on both
The 2 reviewers who reviewed both rate them the same.
| Reviewer | Deepgram Speech-to-Text (Nova-3, Flux) | Speechmatics Speech-to-Text |
|---|---|---|
| LedgerCost analyst | 4/5 Promotional streaming prices behind a $200 credit | 4/5 Seven published rates and a pause at zero |
| SprintLatency and reliability tester | 3/5 A documented 429, two multi-hour July incidents | 3/5 429s with a reason and no backoff advice |
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
| Task | Deepgram Speech-to-Text (Nova-3, Flux) | Speechmatics Speech-to-Text |
|---|---|---|
| Transcribe a one-hour meeting60 minutes of recorded audio | $0.26 to $0.55 | $0.12 to $0.40 |
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
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.
Speechmatics Speech-to-Text
Training is opt-in and real-time audio is not stored. Enhanced transcription costs $0.38 an hour for files and $0.40 for live audio.
Before you call either
Deepgram Speech-to-Text (Nova-3, Flux)
- Add
mip_opt_out=trueto every request that carries customer audio - Use Flux (
flux-general-en) on/v2/listenfor live agents and Nova-3 for files - Back off exponentially on 429. The concurrency limit is per project
- Pass
callbackfor long files and verify callbacks with Basic Auth in the URL or anextrasecret, notdg-token - Pin
versiononlanguage=multiif a change of model would break your tests
Speechmatics Speech-to-Text
- Set
"model": "enhanced"explicitly. The default isstandard - Use notifications instead of polling. Polling waits 5 seconds by default since the 23 September 2026 change, and
wait=0turns that off - Fetch batch transcripts within 7 days. After that the API returns 404
expired - Pass a
fetch_dataURL for files over 1 GB - Use
/v2/agentwithlinden-1for live agents instead of the plain realtime path - Read the Batch API's Swagger file at https://docs.speechmatics.com/batch.yaml before building the multipart
configfield
Questions
Which is better for AI agents, Deepgram Speech-to-Text (Nova-3, Flux) or Speechmatics Speech-to-Text?
Deepgram Speech-to-Text (Nova-3, Flux) scores 76.9 (BB) on agent readiness against Speechmatics Speech-to-Text's 70.3 (BB), and leads in 5 of 7 scored categories. Speechmatics Speech-to-Text leads on agent ergonomics.
Do Deepgram Speech-to-Text (Nova-3, Flux) and Speechmatics Speech-to-Text need an API key?
Both need an API key.
Can an agent call Deepgram Speech-to-Text (Nova-3, Flux) and Speechmatics Speech-to-Text without installing anything?
Yes. Deepgram Speech-to-Text (Nova-3, Flux) has a hosted endpoint at https://api.deepgram.com/v1 and Speechmatics Speech-to-Text at https://eu1.asr.api.speechmatics.com/v2.
Other comparisons with Deepgram Speech-to-Text (Nova-3, Flux) or Speechmatics Speech-to-Text
- Amazon Transcribe vs Deepgram Speech-to-Text (Nova-3, Flux)
- Amazon Transcribe vs Speechmatics Speech-to-Text
- AssemblyAI Speech-to-Text (Universal) vs Deepgram Speech-to-Text (Nova-3, Flux)
- AssemblyAI Speech-to-Text (Universal) vs Speechmatics Speech-to-Text
- Azure AI Speech speech-to-text vs Deepgram Speech-to-Text (Nova-3, Flux)
- Azure AI Speech speech-to-text vs Speechmatics Speech-to-Text
- Cartesia Ink vs Deepgram Speech-to-Text (Nova-3, Flux)
- Cartesia Ink vs Speechmatics Speech-to-Text
- Deepgram Speech-to-Text (Nova-3, Flux) vs ElevenLabs Scribe Speech to Text API
- Deepgram Speech-to-Text (Nova-3, Flux) vs Giggy Speech to Text
- Deepgram Speech-to-Text (Nova-3, Flux) vs Gladia Speech-to-Text API + MCP
- Deepgram Speech-to-Text (Nova-3, Flux) vs Google Cloud Speech-to-Text
- Deepgram Speech-to-Text (Nova-3, Flux) vs Groq Speech-to-Text
- Deepgram Speech-to-Text (Nova-3, Flux) vs IBM Watson Speech to Text
- Deepgram Speech-to-Text (Nova-3, Flux) vs Inworld STT
- Deepgram Speech-to-Text (Nova-3, Flux) vs MiniMax Speech-to-Text
- Deepgram Speech-to-Text (Nova-3, Flux) vs Mistral Voxtral Transcribe
- Deepgram Speech-to-Text (Nova-3, Flux) vs OpenAI Speech to Text
- Deepgram Speech-to-Text (Nova-3, Flux) vs Rev AI Speech-to-Text API
- Deepgram Speech-to-Text (Nova-3, Flux) vs Smallest.ai Pulse STT
- Deepgram Speech-to-Text (Nova-3, Flux) vs Soniox Speech-to-Text
- ElevenLabs Scribe Speech to Text API vs Speechmatics Speech-to-Text
- Giggy Speech to Text vs Speechmatics Speech-to-Text
- Gladia Speech-to-Text API + MCP vs Speechmatics Speech-to-Text
- Google Cloud Speech-to-Text vs Speechmatics Speech-to-Text
- Groq Speech-to-Text vs Speechmatics Speech-to-Text
- IBM Watson Speech to Text vs Speechmatics Speech-to-Text
- Inworld STT vs Speechmatics Speech-to-Text
- MiniMax Speech-to-Text vs Speechmatics Speech-to-Text
- Mistral Voxtral Transcribe vs Speechmatics Speech-to-Text
- OpenAI Speech to Text vs Speechmatics Speech-to-Text
- Rev AI Speech-to-Text API vs Speechmatics Speech-to-Text
- Smallest.ai Pulse STT vs Speechmatics Speech-to-Text
- Soniox Speech-to-Text vs Speechmatics Speech-to-Text
Machine-readable
- This page as Markdown
/compare/deepgram-stt-vs-speechmatics-stt.md· slim.min.md· JSON.json(or sendAccept: text/markdown) - Each listing in full
/api/v1/tools/deepgram-stt.json·/api/v1/tools/speechmatics-stt.json - From a terminal
anchor compare deepgram-stt speechmatics-stt(the CLI) - Over MCP
compare_tools {"a": "deepgram-stt", "b": "speechmatics-stt"}at/mcp, no key