Speechmatics

What it does

Speechmatics is a speech-to-text API that converts recorded audio or video files and live audio streams into transcripts. It also offers a Text to Speech API for generating spoken audio from text, and an Agent STT model that produces turn-based transcription for voice agents.

Developers use it to add captions, transcripts, call analytics, meeting notes, or voice agent inputs to their products. A single model covers 55+ languages, including speakers who switch languages mid-sentence, and speaker diarization and custom vocabulary are included. Deployment can be cloud-hosted, on-premises as containers or a virtual appliance, on GPU or Kubernetes infrastructure, or on-device.

It suits developers and teams in regulated sectors such as healthcare, legal, and media that need compliance certifications. Self-serve accounts start with $100 in credit and no card required, with lower concurrency and file limits than Enterprise, where on-device and on-prem deployment are available.

What it can do

• Real-time speech-to-text transcription from live audio
• Batch transcription of recorded audio and video files
• Multilingual transcription across 55+ languages with code-switching
• Speaker diarization and custom dictionary support
• Text to Speech API for generating speech from text
• Agent STT with turn detection for voice agent pipelines

Before you choose

• Real-time and pre-recorded rates differ by model, and Melia 1, Enhanced, Oak 1, and Standard have different feature sets
• Self-serve plans have lower concurrency and file-per-second limits than Enterprise
• Pre-recorded data auto-deletes after 7 days via the API unless otherwise configured

Listed under

• Coding
• Contact center
• Developers
• Enterprise
• Live captions
• Medical transcription
• Meeting notes

Get started

New accounts receive $100 in credit with no card required; API keys are created in the portal.

1. Create an account at the portal and generate an API key
2. Install the SDK for your language, such as pip install speechmatics-rt for Python
3. Send audio to the Realtime or Batch endpoint and read the returned transcript

Usage-based pricing: 1 credit equals $1 before discounts, consumed per second of audio at hourly rates. Starting credit is time-limited (sign-up bonus credits last 12 months). Check the Speechmatics pricing page for the full rate table and add-on rates.

Source: speechmatics.com/pricing, checked 29 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Speechmatics | Speech APIs powering Voice AI ↗www.speechmatics.com
• Speechmatics pricing: Speech to Text from $0.24/hr ↗www.speechmatics.com
• Welcome | Speechmatics Docs ↗docs.speechmatics.com

Company website

speechmatics.com: product, pricing and policy pages

YC directory mirror

Batch and company name

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.