What does Speechmatics do?

Speechmatics provides speech technology and Voice AI for enterprises, offering accurate Speech-to-Text, Text-to-Speech, and Voice Agent solutions. Our models understand every voice and accent across 55+ languages, helping businesses unlock the full potential of voice data.

How accurate is Speechmatics Speech-to-Text?

Speechmatics delivers best-in-market accuracy, achieving up to 99% word accuracy and 96% medical keyword recall in industry benchmarks. Our models handle multiple accents, noisy environments, and multi speakers with ease.

What makes Speechmatics Text-to-Speech different?

Our low-latency Text-to-Speech (TTS) delivers lifelike, human-sounding voices with sub-150ms latency that is ideal for real-time conversations. Developers can stream natural speech in multiple voices and deploy it in the cloud, hybrid, or on-prem for privacy and control.

Can I build real-time voice agents with Speechmatics?

Our voice AI enables developers to build real-time voice agents that listen, understand, and respond naturally. Plug in fast with a flexible API and native integrations to power your AI voice agents.

Which industries use Speechmatics?

Speechmatics is trusted by organizations in media, healthcare, contact center, medical, finance, legal, education, and accessibility. Our technology powers transcription, translation, call analytics, and voice AI applications worldwide.

Automatic speech-to-text service without the hassle

Why are speech-to-text services hard to use?

I came across a blog from Luis Quintanilla where he produced a console app to transcribe podcasts using the Microsoft Speech API.

The thing that instantly struck me was the hoops that he had to jump though to transcribe the podcast. The app had to:

Transcode the audio to wav as Microsoft don’t support MP3 files
Break the audio up into small chunks so it could be processed
Collate the results and build back into a single transcript

Why, why, why? Every one of those steps is a chance for something to go wrong!

This makes a whole load of work to produce a stable, reliable process to make use of a utility service to get transcripts from audio. After all its called automatic speech recognition so you need it to be automatable reliably right?

Why does it have to be that complex?

Good news… It doesn’t.

Speechmatics believe that it’s hard work and frustrating when using a utility becomes the hard part of a task, rather than being a tool that simply helps you solve the problems. So, the Speechmatics API has been designed to make it easy to consume, not get in the way. With the Speechmatics API you can:

Use the original file (we support just about every file format you can think of)
Support the file upload in one chunk (up to 2 hours or 4GB file size)
If you speak to us nicely we will even let you point directly at the URL of the file you want to transcribe and we will pull it for you so you don’t even have to download it.

What does that mean?

Firstly, why bother using an API at all, just register for using our web app and use the app to transcribe your audio directly and download the result or get the transcript emailed back to you.

Or use the API key after you register and use a single HTTP Post command to request the transcript and a single HTTP to retrieve the result (API docs are available).

Why not give it a go? If you register, we will give you 60-minutes free transcription.

Other services that sound simple can be hard to use so you might not even get to the transcript. Ours is so easy to use and I am sure that the results will be great too.

Ian Firth, Speechmatics