Speechmatics
Speechmatics APIs transcribe live and recorded speech in 55+ languages, returning text for products such as voice agents, captions, and clinical documentation. Developers and regulated-industry teams can deploy the service in the cloud, on-premises, or on-device.
Who it's for
- Voice-agent developers who need low-latency, speaker-aware speech recognition across multiple languages.
- MedTech and healthcare teams building clinical dictation or ambient scribing workflows with the Medical Model.
- Media, legal, and contact-center teams that need transcription for live captions, proceedings, or interaction analytics.
How it fits your workflow
Create an API key, connect it to an application or workflow, and send recorded audio or live audio for transcription. The API returns transcripts that can be used for captions, search, analytics, records, and voice AI.
For real-time Python transcription, install the package, configure the language and audio format, and send microphone audio frames through an API session. The example handles partial and final transcript messages; the service can be deployed in the cloud, on-premises, or on-device.
Pricing
Free: New accounts receive $100 in credit with no credit card required; transcription starts at $0.12 per hour.
Free
$100 in credit; no credit card required. Transcription starts at $0.12 per hour across 55+ languages.
- · Test transcription quality, language support, and integration options
Pro
Credit packs from $25 with no recurring commitment, or monthly subscriptions for predictable spend.
- · Credit packs or monthly subscription options
Enterprise
Custom pricing.
- · For large-scale, regulated, or sensitive deployments
- · Unlimited files per second
- · Real-time concurrency that adjusts to demand
- · Cloud, on-premises, or on-device deployment
Prices checked on Nov 3, 2026 from the vendor's site. They can change; confirm before you buy.
Key features
- Real-time transcription
- Transcribes live speech with partial and final transcript results; the page advertises real-time speech-to-text in 350ms and transcription in less than one second.
- Recorded-file transcription
- Processes recorded audio files and returns transcripts for uses such as captions, search, analytics, and records.
- Multilingual speech recognition
- Supports 55+ languages, including conversations where speakers switch between languages mid-sentence.
- Speaker diarization
- Identifies and separates speakers in conversations, including meetings, calls, legal proceedings, and healthcare consultations.
- Custom vocabulary
- Supports custom vocabulary to help capture terminology accurately.
- Deployment options
- Can run in the cloud, on-premises, or on-device, allowing deployments to match stated privacy and workflow needs.
Getting started
pip install speechmatics-rt- Create an account and API key; new accounts receive $100 in credit with no card required.
- Set the `SPEECHMATICS_API_KEY` environment variable for the supplied Python example.
- Configure the transcription language and audio format, then start a session and send audio frames to receive transcript results.
Frequently asked questions
What is a speech-to-text API and how does Speechmatics use one?
A speech-to-text API converts spoken audio into written text inside an app, product, or workflow. Speechmatics can process recorded files or live audio and return transcripts for captions, search, analytics, records, and voice AI.
Can I try Speechmatics for free?
Yes. New accounts receive $100 in credit with no card required, so users can test transcription quality and language support.
How many languages does Speechmatics support?
The page says Speechmatics supports 55+ languages, including accents, dialects, and code-switching within a conversation.
Can Speechmatics distinguish speakers?
Yes. Speaker diarization identifies and separates speakers in meetings, calls, legal proceedings, and healthcare consultations.
Is Speechmatics suitable for regulated industries?
The page says Speechmatics supports cloud, on-premises, and on-device deployment and lists ISO 27001, GDPR, HIPAA, and SOC 2 Type II compliance.