Speechmatics

What is Speechmatics?

Speechmatics is a UK-founded speech recognition company selling a developer-facing API for transcribing audio and video into text, in both real-time (streaming) and batch modes, across 55+ languages. It offers three transcription models: Standard (fast), Enhanced (highest accuracy), and Melia (multilingual, handling speakers who switch languages mid-conversation without needing a preset language pack).

The API supports speaker and channel identification, custom dictionaries for jargon, confidence scores, and deployment via cloud, on-premises, or on-device for security-sensitive customers. It's positioned for companies building voice agents, transcription products, call-center analytics, or captioning at scale, rather than as a consumer app end users interact with directly.

Key features

• Real-time (WebSocket) and batch transcription APIs across 55+ languages
• Melia model transcribes multilingual/code-switching audio without preset language packs
• Standard and Enhanced models trade off speed versus best-in-class accuracy
• Speaker and channel identification with custom dictionaries for jargon
• On-prem, cloud, or on-device deployment for regulated industries
• Volume-based discounts and an opt-in data-sharing discount on usage

How to get started

1. Sign up for a free account and receive $100 in usage credit
2. Choose real-time or batch transcription and pick a model
3. Send audio to the REST API or the WebSocket streaming endpoint
4. Configure language, speaker diarization, and custom dictionary settings
5. Add a payment card to continue on pay-as-you-go pricing after free credit

Speechmatics pricing

New accounts get $100 in free credit with no card required. Beyond that, the self-serve Pro plan bills pay-as-you-go from $0.129/hour, with automatic volume discounts above 500 hours/month and an opt-in data-sharing discount; Enterprise is billed on a custom contract basis.

Our take: Speechmatics' Melia model transcribing code-switched, multilingual audio into one continuous transcript without requiring a preset language pack is a fairly unusual capability among speech-to-text APIs.

Prices verified 2026-09. Plans change often — confirm on the official site above before you buy.

Use cases

• Developers building voice-agent or call-center transcription products
• Companies needing accurate captions or subtitles at scale
• Teams transcribing multilingual meetings where speakers switch languages
• Enterprises requiring on-prem or on-device speech-to-text for compliance

Speechmatics vs. alternatives

FAQ

Is Speechmatics free to use?

Speechmatics's plans: $100 free credit, no card required; Pro pay-as-you-go from $0.129/hr; custom Enterprise. Pricing and free-tier limits change over time, so check the official site above for the latest details.

What is Speechmatics used for?

Enterprise speech-to-text API for real-time and batch transcription across 55+ languages, including code-switching audio.

Who is Speechmatics best for?

Speechmatics is a good fit for developers building voice-agent or call-center transcription products, or companies needing accurate captions or subtitles at scale.

What are the alternatives to Speechmatics?

Commonly compared alternatives include AssemblyAI, RambleFix, and Otter.ai, see the comparison above for how they differ.

How much does Speechmatics cost?

Speechmatics gives new accounts $100 in free credit with no card required. Beyond that, Pro usage is pay-as-you-go starting from $0.129/hour, and Enterprise deployments are custom-quoted.