Speechmatics
Speechmatics is a UK-based enterprise speech-to-text API with strong accent and dialect coverage across 55+ languages, a managed cloud, and a fully supported on-prem container deployment for regulated workloads.
Drop your audio. Transcript in seconds. First transcript free, then $2 a file or $8 = 1,000 min
Speechmatics offers batch and real-time speech-to-text across 55+ languages, with multilingual packs (e.g. Mandarin-English, Spanish-English), Standard and Enhanced operating points, plus a separate medical model for clinical audio. The API surface is one HTTPS endpoint for batch (POST /jobs) and a WebSocket for streaming, with a managed cloud in EU / US / AU regions and a supported on-prem container path for environments that cannot send audio to a third-party cloud.
Best for regulated enterprise buyers — broadcasters, contact centers, legal / public-sector, and healthcare — that need strong recognition across heavy accents, a sovereign or self-hosted deployment option, and the procurement paperwork a major vendor provides. The Free tier gives 480 minutes/month of speech-to-text and 2 concurrent real-time sessions, no card. Pro is usage-based from $0.24/hr with the same 480 min/month included, 50 concurrent real-time sessions, and 10 file jobs/sec. Enterprise is custom-priced with volume discounts above 500 hours/month, unlimited concurrency, and on-prem options. Last price check: 2026-05-10.
What it is
Speechmatics is the enterprise incumbent — strong on heavily-accented English, full on-prem deployment, and the compliance paperwork big buyers require. Not price-competitive for indie projects, but often the only viable option for a regulated enterprise buyer. Last price check: 2026-04-20.
Watch out for: Pricing is quote-based and typically higher than self-service APIs.
Install / use
Where Speechmatics fits · 6 segments
Speechmatics is the enterprise incumbent — pick the card closest to your workload. Each links to the canonical product or docs page.
Transcribe call recordings or live conversations for QA scoring, agent assist, and compliance review. The Enhanced operating point handles heavy phone-line accents and noisy audio better than Standard, and diarization with word timestamps lets you label agent vs caller turns for downstream analytics.
Batch and real-time both supported
Used by broadcasters and media-monitoring vendors for live captioning, post-production transcripts, and back-catalogue indexing. Wide accent and dialect coverage in a single English pack means you do not have to pre-route audio to a regional model.
Single English model covers global accents
Hearings, depositions, and discovery audio batched through the jobs API with diarization and word timestamps for downstream verbatim editing. Procurement-friendly contract terms and a documented retention policy make this a common pick for law firms and court-reporting vendors.
Word-level timestamps for verbatim editing
Government and public-sector buyers that cannot send audio to a US-based cloud can pick a regional SaaS endpoint (EU / US / AU) or the on-prem container path. Same model behind both, with the same accent and language coverage.
Regional data residency
Speechmatics positions strong accent coverage inside a single English pack — Indian, African, Caribbean, regional UK and US accents are all handled by the same model id, so you do not pre-classify the speaker. Useful for global call centers, ed-tech, and consumer apps with international users.
One English model for all accents
Speechmatics ships CPU and GPU containers for batch and real-time, plus Kubernetes manifests and a separate language-identification container. Maximum control over data and deployment for HIPAA, financial, and defense workloads where audio cannot leave the customer environment.
Same model behind SaaS and on-prem
Quickstart · pick a runtime
Three working ways to talk to Speechmatics. Export your key as SPEECHMATICS_API_KEY first — grab one from the Speechmatics portal after signup (free 480 min/month, no card).
Official speechmatics-batch package · transcribe a local audio file end-to-end.
# pip install speechmatics-batch python-dotenv
import asyncio
import os
from speechmatics.batch import AsyncClient
async def main():
client = AsyncClient(api_key=os.environ["SPEECHMATICS_API_KEY"])
result = await client.transcribe("audio.wav")
print(result.transcript_text)
await client.close()
asyncio.run(main())Plain HTTPS POST to /jobs · useful for shell pipelines and CI runners.
# submit a batch job by URL
curl -X POST 'https://api.speechmatics.com/jobs' \
-H "Authorization: Bearer $SPEECHMATICS_API_KEY" \
-H 'Content-Type: application/json' \
-d '{
"config": {
"type": "transcription",
"transcription_config": {
"language": "en",
"operating_point": "enhanced",
"diarization": "speaker"
}
},
"data_uri": "https://example.com/audio.wav"
}'Streaming session via the SDK · the AsyncClient manages the WebSocket connection for you.
# pip install speechmatics-rt python-dotenv
import asyncio
import os
from speechmatics.rt import AsyncClient, TranscriptionConfig, AudioFormat
async def main():
config = TranscriptionConfig(language="en", operating_point="enhanced")
audio_format = AudioFormat(type="raw", encoding="pcm_s16le", sample_rate=16000)
async with AsyncClient(api_key=os.environ["SPEECHMATICS_API_KEY"]) as client:
await client.start_session(
transcription_config=config,
audio_format=audio_format,
)
# ... feed PCM chunks via client.send_audio(...) and read transcripts ...
asyncio.run(main())What people actually do with Speechmatics-style transcription
The tool is the means. These are the jobs — each one priced at published rates, each one wired up on its own page.
Features
| Speaker diarization | Yes |
| Word-level timestamps | Yes |
| Streaming / real-time | Yes |
| Languages supported | 50 |
| HIPAA eligible | Yes |
Links
- speechmatics.com ↗Product homepage — speech-to-text, voice agents, text-to-speech, and the 55+ language pitch.
- speechmatics.com/pricing ↗Current plan tiers — Free (480 min/month), Pro from $0.24/hr, Enterprise with volume discounts above 500 hrs/month.
- docs.speechmatics.com ↗Documentation root — quickstarts, batch and real-time guides, deployments, and API reference.
- API reference — create a batch job ↗POST /jobs reference with config schema, diarization, operating points, and translation options.
- Supported languages ↗Full language table — 55+ languages with Standard and Enhanced operating points, plus bilingual packs.
- Deployments — SaaS and on-prem ↗Comparison of SaaS regions (EU / US / AU) vs on-prem containers (CPU, GPU, Kubernetes, language-id).
- github.com/speechmatics ↗Official GitHub organization — Python SDK, JS / TS SDK, CLI, examples, and community repos.
- speechmatics/speechmatics-python-sdk ↗Official Python SDK — async clients for batch and real-time transcription.
- speechmatics/speechmatics-js-sdk ↗Official JavaScript / TypeScript SDK — browser, Node 18+, and edge runtimes.
- status.speechmatics.com ↗Live status for Batch SaaS (EU / US / AU), Realtime SaaS (EU / US), portal, docs, and on-prem registry.
- portal.speechmatics.com ↗Self-service portal — create API keys, view usage, manage billing; signup gives 480 min/month free, no card.
- speechmatics.com/blog ↗Product and engineering blog — recent posts on alphanumeric recognition, on-device deployments, and STT comparisons.
Speechmatics vs Whipscribe
| Feature | Speechmatics | Whipscribe |
|---|---|---|
| Category | Transcription APIs | Transcription APIs |
| Pricing | Not verified | $8–$24 one-time packs (credits never expire) · $2 single unlock · free instant preview |
| Speaker diarization | Not verified | Yes |
| Word timestamps | Not verified | Yes |
| Streaming | Not verified | No |
| Languages | 50 | 99 |
| Platforms | API, On-prem | Web, API, MCP |
Where this category is heading
From the vendor changelogs we track weekly — what changed in August 2026, and what it means if you are choosing now.
AssemblyAI moved summarisation onto an LLM this month; every vendor is racing to return action items, quotes and topics with the text rather than as an add-on.
Whipscribe today Every Whipscribe job already returns an insights payload — summary, key quotes, topics and speakers — from the same job id, at no extra charge.
Deepgram shipped self-hosted container images in August — the market is moving toward audio that stays inside a boundary the customer controls, because teams with customer calls or unreleased material are refusing shared model endpoints.
Whipscribe today Whipscribe runs on our own GPUs in a private, secured cloud. Audio is never forwarded to OpenAI or any third-party model.
The fastest-growing way to use a transcription API is not a form — it is Claude, Cursor or a workflow runner calling it mid-task through MCP.
Whipscribe today Whipscribe ships an MCP server: transcribe, search and summarise from an assistant without wiring anything.
AssemblyAI's 1.0 SDK unified async, realtime and sync; Deepgram's CLI went to 0.3. The unit of work is becoming the folder or the bucket, not the file.
Whipscribe today Submit with an Idempotency-Key and a batch_id, poll by job, retry safely. The S3 connector runs a whole prefix in one grant.
Deepgram added Afrikaans, Georgian and Armenian and improved a dozen more this month. Coverage is widening while quality still clusters around English and the large European languages.
Whipscribe today 99+ languages auto-detected. Ask for a language explicitly when you know it — auto-detect on a short or noisy clip is the most common cause of a wrong-language transcript.
Source: Deepgram and AssemblyAI changelogs, scanned 2026-08-24.
Alternatives to Speechmatics
Frequently asked about Speechmatics
What is Speechmatics?
Speechmatics offers batch and real-time speech-to-text across 55+ languages, with multilingual packs (e.g. Mandarin-English, Spanish-English), Standard and Enhanced operating points, plus a separate medical model for clinical audio. The API surface is one HTTPS endpoint for batch (POST /jobs) and a WebSocket for streaming, with a managed cloud in EU / US / AU regions and a supported on-prem container path for environments that cannot send audio to a third-party cloud.
How much does Speechmatics cost?
Speechmatics does not publish a self-serve price — it is quoted by the vendor on contact. Expect an enterprise-style agreement rather than a card-and-go signup.
Does Speechmatics work on Mac, Windows and Linux?
Speechmatics runs on API and On-prem. It is not listed for other platforms; check the vendor for anything beyond API and On-prem.
How do I get started with Speechmatics?
Speechmatics is used through its API: get a key from the vendor, then call it from your own code. There is no desktop app to install.
How many languages does Speechmatics support?
Speechmatics lists 50 languages.
What are the limitations of Speechmatics?
Pricing is quote-based and typically higher than self-service APIs.
Who is Speechmatics best for?
Regulated enterprise (banks, broadcasters, public sector) needing on-prem or sovereign deployment.
What should I know before choosing Speechmatics?
Best for regulated enterprise buyers — broadcasters, contact centers, legal / public-sector, and healthcare — that need strong recognition across heavy accents, a sovereign or self-hosted deployment option, and the procurement paperwork a major vendor provides. The Free tier gives 480 minutes/month of speech-to-text and 2 concurrent real-time sessions, no card. Pro is usage-based from $0.24/hr with the same 480 min/month included, 50 concurrent real-time sessions, and 10 file jobs/sec. Enterprise is custom-priced with volume discounts above 500 hours/month, unlimited concurrency, and on-prem options. Last price check: 2026-05-10.
Is Speechmatics open source?
No. Speechmatics is a proprietary API. If you need source you can read and run yourself, the open-source tools in this directory are the place to look.
Can I use Speechmatics in the browser?
No — Speechmatics is a API and On-prem application, not a web app. You install it rather than sign in to it.
What kind of tool is Speechmatics?
In this directory Speechmatics is filed under commercial api, enterprise as a API.
What are the alternatives to Speechmatics?
There is a side-by-side page at /tools/speechmatics-alternatives comparing Speechmatics with the closest tools in the same category on price, platform and features.
Are there setup recipes for Speechmatics?
Yes — this page carries 3 tested setups: Python SDK · batch transcription; cURL · submit a batch job; Python · real-time WebSocket.
Can I automate Speechmatics-style transcription without running it myself?
Yes. If what you want is transcripts rather than the tool itself, Whipscribe does the same job as a hosted API: submit a file or URL, poll a job id, pull the result as txt, json, srt, vtt or docx. $0.008 a minute — about $0.48 an audio hour — bought as credits that never expire. The automation recipes on this site show it wired to Drive folders, Zoom recordings, S3 buckets and no-code platforms.
How is Whipscribe different from Speechmatics?
Speechmatics is another vendor's product; Whipscribe is a per-minute transcription service. We run the models on our own GPUs in a private cloud — audio is never forwarded to OpenAI or any third-party model — and every job returns five formats plus an AI summary, quotes and topics. Your first transcript in the web app is free at any length, no card.
Whipscribe is a managed faster-whisper + whisperX service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.
Explore
All transcription tools · Audio technology hub · Transcribe any platform · Audio & video formats · How-to guides · Glossary · Playbooks · Apps · Broadcast & radio · Podcast transcripts · Use cases · Blog · Transcription API · Integrations · Automations · For your industry