Speechmatics Logo New

Editorial team

Speechmatics, the leading speech recognition technology scaleup, unveils plans to amplify its real-time transcription capabilities by providing real-time speech translation in an all-in-one API.

Real-Time Translation

Speechmatics to launch pioneering real-time speech translation capabilities in 69 language pairs 

Company

Bethan Thomas author pic

Senior Machine Learning Engineer

Our latest Ursa release was able to achieve incredible accuracy partly through scaling self-supervised learning. In this blog we demonstrate the power of self-supervised learning and challenge the assumption that scaling labeled data is the key to greater accuracy. We show that with 300x less the amount of labeled data we still beat the nearest vendor by 12% relative.

Self-Supervised Learning

Boosting sample efficiency through Self-Supervised Learning

Technical

Ana Olssen

Machine Learning Engineer

Speechmatics' newest model, Ursa, performs better than ever on specialized topics like medicine and finance.

Join Ana Olssen as she explains how we can understand every voice in specialized domains by relying on self supervised learning combined with powerful language models.

Domain Generalization image

Ursa: Scaling up as a Solution to Domain Generalization

Lawrence Atkins

David MacLeod

Machine Learning Architect

In a world of increasingly costly machine learning model deployments, ensuring accurate GPU operation timing is key to resource optimization. In this blog post, we explore best practices to achieve this in PyTorch.

PyTorch timing operations

How to Accurately Time CUDA Kernels in Pytorch

By combining the output of an automatic speech recognition (ASR) system with a large language model such as GPT-4 and ChatGPT, we can do many downstream tasks in a zero-shot way by including the transcript and task in the prompt.

GPT-4 - What You Need To Know

Accuracy Matters When Using GPT-4 and ChatGPT for Downstream Tasks

John Hughes

Accuracy Team Lead

On March 14 OpenAI released GPT-4, a multi-modal large language model that can accept both images and text as input. Learn how GPT-4 works and what it means in this technical blog.

GPT4 - What is it?

GPT-4: How does it work?

Owen O-Loan

Director of Engineering Operations

Senior Product Manager, Owen O’Loan, introduces the world to Speechmatics’ latest product release, including the first look at our new Translation offering.

Accuracy Uplift blog header

Product Release March 2023: Uplifts in Accuracy, Translation, and Automatic Language Identification

Product

Benedetta Cevoli

Our latest release, Ursa, breaks the accessibility barriers in speech technologies by offering ground-breaking accuracy for every voice. Irrespective of accent, dialect, and various demographic factors, Ursa is consistently the most accurate speech-to-text engine on the market with a relative lead of up to 30% compared to Amazon, Google, Microsoft, and OpenAI’s Whisper across the board.

Accessibility blog header image

Achieving Accessibility Through Incredible Accuracy with Ursa

Will Williams

Chief Technology Officer

We are excited to introduce Ursa, the world’s most accurate speech-to-text system. Ursa delivers unprecedented performance across a diverse range of voices. We observe relative accuracy gains of 22% and 25%* versus Microsoft and OpenAI’s Whisper respectively.

Ursa provides the world’s most accurate speech-to-text and is a breakthrough in accessibility, reducing the digital divide for voices that other speech-to-text systems struggle to recognize.

Ursa - header image

Introducing Ursa from Speechmatics

Speechmatics and Ubisoft partner to offer real-time captioning to significantly boost accessibility in live stream gaming 

Speechmatics and Ubisoft partner to add accessibility to live stream gaming

Accessibility for Gaming

Ubisoft-owned game Brawlhalla pushes industry benchmark on accessibility with Speechmatics

From ultra-fast conversational AI to studio-quality narration, find the voice that matches your use case and budget.

Best TTS APIs in 2026: Top 12 Text-to-Speech services for developers

Use Cases

Sign In

Get started

Contact sales

Speech-to-Text API

Capabilities

Text-to-Speech API

Voice Agents

Speech-to-text

contact center icon

Contact Center Solutions

Medical and healthcare-Navigation icon

Medical & Healthcare

media-and-captioning-icon

Media Distribution & Captioning

technical-blog-icon

AI Voice Agents

meeting-platforms

Note-Taking & Meeting Assistants

media-monitoring-icon

Speech Analytics

market-exposure-icon

Status page

By Use Case

blog-icon

News & Articles

events-icon

Partnerships

contact-us-icon

Contact Us

careers-icon

Careers

about-speechmatics-icon

Find Out More About Speechmatics

case-studies-icon

Case Studies

Resources

Technical information

documentation-icon

service-status-icon

Service Status

Discord icon

Join the community

Technical blog

announcement-icon

GitHub Academy

Startup Program

Developers

Enterprise

Pricing

Advanced speech recognition API with high accuracy across [<langcount>] languages.

Speechmatics Speech-to-Text

Get started with our speech recognition API

Free

Speechmatics

Advanced features for professional use ($0.24 per hour)

Custom solutions for large-scale deployments

We are Speechmatics. Global experts in deep learning and voice technology. Accurate speech-to-text for analysis and accessibility.

Speechmatics Social Sharing

Global experts in deep learning and voice technology. Accurate speech-to-text for analysis and accessibility.

Stay up-to-date with the very latest product news, company news and artificial intelligence (AI) research. Bookmark our technology blog today!

Blog & Latest Speech Recognition News| Speechmatics

Word error rate for legal transcription has no single acceptable threshold. But knowing how accuracy, audio quality, and review obligations connect to real legal risk is what separates a reliable transcript from a costly one.

What Word Error Rate Is Acceptable for Legal Transcription?

The court reporter shortage is reshaping litigation. Explore data, causes, and how legal teams are using digital reporting and AI transcription to adapt.

The court reporter shortage crisis: data, causes, and what legal teams are doing about it

Why predicting durations as well as tokens allows transducer models to skip frames and achieve up to 2.82X faster inference.

How Nvidia Dominates the HuggingFace Leaderboards in This Key Metric

AI-powered Arabic medical transcription for hospitals and clinics across the Middle East, improving accuracy, compliance, and speed.

Speechmatics Startup Program: Unlock $25K Credits

Latest

Turning real-time clinical speech into trusted, EHR-native automation.

One word changes everything: Speechmatics and Edvak EHR partner to make voice AI safe for clinical automation at scale

The metrics, milestones, and real-world outcomes that defined Voice AI’s breakout year.

Speechmatics in 2025: the numbers that shaped Voice AI's breakthrough year (+ what’s to come in 2026) 

Ship Voice AI agents that understand every voice in real-time, even in noisy, multi-speaker scenarios.

Vapi and Speechmatics: Build agents that understand every voice

Real-time voice AI isn’t just about speed — it’s about handling thousands of live conversations at once. Here’s why concurrency makes or breaks performance at scale.

The scaling challenge voice AI can’t ignore

Expanding a Nordic medical lineup with 3.91% KWER model that delivers sub-second latency across Swedish, Finnish, Danish, and Norwegian clinical workflows.

Speechmatics launches new Swedish medical model, cutting transcription errors by 40%

Built on NVIDIA AI infrastructure, the collaboration combines medical-grade speech models and agent workflows to deliver measurable ROI; cutting admin time and scaling healthcare capacity worldwide.

Speechmatics and Sully.ai partner to scale healthcare AI infrastructure globally

The voice AI gold rush has made natural-sounding speech a premium commodity, but most businesses don’t need to pay Oscar-winning prices to sound human.

Do you really need to pay a fortune for your text-to-speech?

New German, Danish and Norwegian Medical Models deliver up to 50% lower error rates, real-time accuracy under one second, and full deployment flexibility.

Speechmatics sets new standard for real-time medical transcription with German and Nordic roll-out

Joining French, Dutch, Finnish and English for global clinical transcription - accurate, hallucination-free, and accent-independent.

Speechmatics Medical Model launches in Spanish

Most TTS sounds great in demos but breaks in real conversations. We built ours for sub-150ms latency, natural voices, and global scale.

Why we built our low-latency Text-to-Speech

50% fewer critical errors. 96% medical term recall. A new benchmark for real-time clinical speech recognition.

The next generation of AI medical speech recognition is here

Next-gen medical transcription model cuts errors, boosts accuracy, and brings real-time reliability to clinical documentation.

Speechmatics sets record in medical Speech-to-Text with 93% accuracy

What “fast” actually means for voice agents — and why Pipecat’s TTFS + semantic accuracy is the clearest benchmark we’ve seen.

Speed you can trust: The STT metrics that matter for voice agents

Introducing 250ms final transcripts for Voice AI

You can’t hurry love, but you can hurry final transcripts

Is it time to rethink responsiveness as a conversational problem, not just a technical one?

Voice AI doesn't need to be faster. It needs to read the room. 

A practical guide to the leading AI agent orchestration platforms shaping real-time Voice AI in 2025, and what to consider before you build.

The 7 best AI agent orchestration platforms for building Voice AI agents in 2025

Benchmarks love milliseconds. 
Users love conversations that actually work.

Why “fastest” voice tech is a trap 

Why supporting thousands of parallel sessions is more than a backend problem.

How we built real-time concurrency for Voice AI at scale

Build real-time, speaker-aware Voice Agents with LiveKit + Speechmatics

Introducing real-time, speaker-aware Voice Agents with LiveKit + Speechmatics

From stage to strategy: core takeaways from a lead engineer on the future of live, usable Voice AI.

Inside the future of Voice AI: Speechmatics at the LiveKit developer showcase 

Open-source TTS is reshaping the landscape — unlocking custom voices, on-device deployment, and fine-tuned speech that fits your domain and your users.

Custom Voice AI in 2025: The Open Source Boom

As clinical workflows become automated and AI-driven, real-time speech is shifting from a transcription feature to the foundational intelligence layer inside modern EHR systems.

Why AI-native EHR platforms will treat speech as core infrastructure in 2026

Voice AI moved fast. Regulation caught up. Here’s how to stay compliant in 2026.

Your essential 2026 guide to voice ai compliance in today's digital landscape

What teams deploying Voice AI at scale see coming next as production pressure reshapes adoption.

7 Voice AI predictions from teams building at scale in 2026

Faceless YouTube channels are becoming a scalable, system-driven way for businesses to produce high-impact video without relying on on-camera talent.