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.

5 Easy Steps to Get Started with Speechmatics

To stay at the summit of the speech-to-text mountain, we recognize that not everyone will immediately know how to use our award-winning Autonomous Speech Recognition engine (ASR). 

Moreover, our mission to understand every voice heavily centers around accessibility within media – for example, open and closed captions are becoming less of an accessory and more of a necessity. So, naturally, we want to keep our highly advanced AI accessible and easy to use. 

In pursuit of that, we've created a helpful guide with five easy steps for getting started with Speechmatics. 

Step 1: Choose Your Deployment Options

Firstly, you will need to choose your deployment options. You have four choices: virtual appliance, containers, SaaS, and hybrid. Here's a short summary of each:

Virtual Appliance

A pre-configured virtual machine capable of doing Real-Time or Batch processing. It can be deployed directly in your on-premises environment.

Containers

Our Docker Containers enable you to build scalable transcription services within your infrastructure in Real-Time or Batch processing.

SaaS

We can deliver all the benefits of the Speechmatics ASR without the complexities of deploying it within your team and environment. Choose public (hosted by Speechmatics) or your cloud.

Hybrid

Hybrid deployment helps those with a mixture of data requirements that use cloud and on-premises processing.

First choice down, four to go. 

Step 2: Choose Your Offering

Next on the guide is to choose your offering. In this instance, we provide you with two options: Batch and Real-Time.

Firstly, there's Batch, where you transcribe speech-to-text from pre-recorded media files at your convenience using our ASR. You can schedule a transcription at a time that suits you.

With the Real-Time offering, you can get your speech-to-text in real time and get results instantly. As a result, you can gather actionable data as soon as needed. If you're worried accuracy would be compromised, fear not; our proprietary technology delivers best-in-class accuracy even at low latencies - proven in recent research against our competitors.

Step 3: Choose Your Features

At this point, your options open some more. Our features range from channel diarization to flexible endpointing. Here's a complete list: 

Entity Formatting
Notifications
Speaker Diarization
Partials
Channel Diarization
Transcript Finalization
Custom Dictionary and Sounds Feature
Flexible Endpointing
Speaker Change

Speech-to-text is not a solved problem yet, so we're always looking to innovate our ASR, expect to see more features in the future. 

Step 4: Choose Your Format

Next up, you will need to choose your format. Happily, we support all major audio and video formats, reducing users' time to prepare files. After all, we're all about speech-to-text accessibility.

For clarity, the default for our output is JSON. Users also have the option to pick an alternative in srt and txt.  

Step 5: Choose Your Language

Despite competition from the most prominent names in the speech-to-text and surrounding AI industry (Google, Microsoft, etc.), we are proud to boast about the 50 languages in our ASR's coverage.

This includes Arabic, Bulgarian, Catalan, Cantonese, Croatian, Czech, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Hungarian, Indonesian, Italian, Japanese, Korean, Latvian, Lithuanian, Malay, Mandarin (traditional and simplified), Norwegian, Persian, Polish, Portuguese, Romanian, Russian, Slovakian, Slovenian, Spanish, Swedish, Turkish, and Ukrainian.

Arabic, Bashkir, Basque, Belarusian, Bulgarian, Cantonese, Catalan, Croatian, Czech, Danish, Dutch, English, Esperanto, Estonian, Finnish, French, Galician, German, Greek, Hindi, Hungarian, Indonesian, Interlingua, Italian, Japanese, Korean, Latvian, Lithuanian, Malay, Mandarin (Traditional & Simplified), Marathi, Mongolian, Norwegian, Polish, Portuguese, Romanian, Russian, Slovak, Slovenian, Spanish, Swedish, Tamil, Thai, Turkish, Ukrainian, Uyghur, Vietnamese and Welsh.

Getting Started Is Only a Few Choices Away

There you have it. This guide was designed to be simple, so with that, here's a short summary:

Choose your deployment options: virtual appliance, containers, SaaS, and hybrid. 
Choose your offering: batch or real-time.
Choose your features: entity formatting, notifications, speaker diarization, partials, channel diarization, transcript finalization, custom dictionary, flexible endpointing, and speaker change. 
Choose your format: all major audio and video formats.
Choose your language: 50 dialects at your disposal.

Paul Gordon, Product Marketing Manager, Speechmatics