What Speechmatics does
Speechmatics provides real-time and batch speech-to-text, text-to-speech, and voice-agent APIs with cloud, private, container, and on-device deployment options.
Speechmatics covers batch and real-time transcription across more than 50 languages, with speaker diarization, word timing, custom vocabulary, language identification, and formatting. It also offers low-latency text-to-speech and voice-agent building blocks. Deployment options range from managed multi-region SaaS to enterprise private, container, virtual-appliance, and eligible on-device configurations.
The official pricing page reviewed July 30 lists a free monthly allowance and Pro speech-to-text rates that vary by model, including a lower-priced batch model, plus separate charges for translation, summaries, chapters, sentiment, topics, and text-to-speech. The page also announces a move to credits on August 1, 2026 without a stated price change. Buyers should confirm the post-transition meter, concurrency, discounts, and enterprise deployment costs.
Speechmatics advertises ISO 27001, SOC 2 Type II, GDPR, and HIPAA-aligned offerings, but compliance depends on the plan, contract, region, and customer configuration. Audio can contain sensitive identifiers and inferred traits. Consent, minimization, retention, access, and deletion remain customer responsibilities, and generated summaries or sentiment should never override the recording or a qualified reviewer.
How Speechmatics works
Developers stream audio or upload a file with language, diarization, vocabulary, formatting, translation, or analysis settings. Speechmatics returns timed text and optional speaker or derived fields; its text-to-speech API turns text into streaming audio, and agent integrations connect these layers to orchestration. Cloud, private, container, or on-device choices determine where eligible processing occurs. Operators must verify transcripts, translations, summaries, and any action triggered from them.
How to set up Speechmatics
Choose processing and deployment
Map languages, real-time or batch needs, cloud region, private or device requirements, concurrency, latency, and regulated data.
Confirm current commercial terms
Verify the August credit transition, model rates, free allowance, bolt-ons, volume discounts, support, and enterprise deployment scope.
Secure the integration
Use server-side keys, least privilege, encryption, regional routing, approved logs, retention, deletion, and participant consent.
Evaluate real-world audio
Test noise, crosstalk, accents, code-switching, names, digits, jargon, diarization, translation, and long streams.
Operate with review
Confirm high-impact fields against audio, expose corrections, sample quality by language, and monitor latency, failures, concurrency, and spend.
Speechmatics FAQs
Is Speechmatics free?
The current pricing page lists a monthly free allowance. Limits and the announced credit transition should be confirmed at signup.
How much is speech-to-text?
Pro rates vary by batch or real-time model, with separate optional analysis charges; consult the live page for the exact post-August meter.
How many languages are supported?
Speechmatics currently advertises more than 50 languages, with exact features and model coverage documented per release.
Can it run outside the public cloud?
Enterprise options include private cloud, container, virtual appliance, and eligible on-device deployments.
Does certification make every workload compliant?
No. Contract, region, access, consent, retention, deletion, and the customer's own controls determine compliance.
Listing reviewed 2026-07-30. Product details and pricing can change; verify important terms on the provider's website.
Related Meetings AI tools
Related AI guides
Reviews
Tell the community what you made, what worked, and what you wish you knew before starting.