Speechmatics now supports Romanian


Speechmatics delivers another new language to the fold, by releasing support for Romanian.

This takes our released languages count now to sixteen, with many more languages either in beta or in development.

One of the many pleasures of working with languages is taking the opportunity to explore the history of the language you are working with, and Romanian as one of the the oldest European languages is no exception. This modern language, that evolved from Vulgar Latin between the sixth and ninth centuries forms a branch of the Italic languages within the Indo-European language family.

Spoken by some 24,000,000 people, we are pleased to now be able to offer the Speechmatics Automatic Speech Recognition system to this population.

Speechmatics deliver Hungarian language capability


Speechmatics deliver another new language to our rapidly growing inventory of language capability.

Hungarian is now the 15th language pack to be delivered using our ground-breaking Auto-Auto language creation framework, proving that Speechmatics has the technology to deliver new language capability quickly and efficiently.

Hungarians call their language “magyar” and its grammar and vocabulary are completely different to the Indo-European languages which surround Hungary. The language is not related to any other Central European language but belongs to the Finno-Ugric language phylum, which includes Finnish and Estonian. Wondering what language we could be delivering next.



Speechmatics release Greek

greekSpeechmatics are pleased to announce the release of our new Greek language capability.

Available for both our public cloud-based service and private on-premise implementations of our world-beating Speech Recognition platform.

Greek being an independent branch of the Indo-European family of languages has the longest documented history of any living language, spanning 34 centuries of written records and holds an important place in the history of Western culture.

Speechmatics welcome everyone to sign-up for a free trial of our public speech transcription service, now available in 14 languages.

Business Weekly also likes the look of speech technology business Speechmatics

Nice write-up…

Business Weekly New Killer 50 whipping up a Storm

Business Weekly also likes the look of speech technology business Speechmatics. Cambridge has proved a honeypot for American technology giants looking for the next big thing in the speech and general AI fields. Microsoft, Amazon, Apple and Google have already collectively shelled out hundreds of million dollars for Cambridge technology in the sector.

Speechmatics is ramping headcount in a rapid scale-up having launched Universal Time Alignment – a language-independent forced-alignment service to match words in text files to their counterparts in audio files. The technology accurately and automatically delivers improved content discoverability, in any language.

The technology was pioneered by Dr Tony Robinson 30 years ago at the University of Cambridge and is now pushing the boundaries in speaker-independent automatic speech recognition.

Speechmatics releases Russian language support


Speechmatics continues to prove our ground-breaking Auto-Auto technology, by releasing yet another language to market. Auto-Autos key ability is to deliver new language models in super-fast time to a very high standard, through the application of the very latest in Machine Learning technology.

Delivering Russian language capability for our Speech Recognition platform allows us to reach another 260,000,000 users, rapidly expanding our global presence.

Deployable on-premise or through our public and private cloud services, allowing a huge range of use-cases to be supported, from banking compliance to robotics and personal virtual assistants.

Contact neilm@speechmatics.com for more information and a free trial.


Speechmatics ready to support MiFID II compliance


Speechmatics is a leader in automatic speech recognition which enables clients to convert speech to text immediately and hence make audio content such as phone calls discoverable. This is also one of the key requirements of the incoming MiFID II regulation, an incoming European wide financial services regulation to improve transparency in the financial services industry. The regulation is set to take effect from January 2018.

 One of the key mechanisms of MiFID II is around telephone taping of financial advisers to support regulatory compliance, protect consumers and to resolve any trading disputes cost effectively. According to IHS Market and Expand, a Boston Consulting Group Company, MiFID II will cost the industry more than $2bn in 2017. The incoming legislation also received further strength and backing through a report published by the Financial Conduct Authority (FCA) last week urging financial institutions to ready themselves.

As a specialist in multilingual, automatic speech recognition, Speechmatics is ideally positioned to help companies that offer turn-key solutions to this upcoming market opportunity by allowing them to convert what was recorded on a call into an accurate transcription. Speechmatics covers most of the core European languages and alongside a cloud-based service, provides a full on-premises solution that easily integrates into any existing software stack, thereby maintaining privacy and security of data.

 Speechmatics is already working with a number of leading call recording and compliance partners in the UK and across Europe.

For more information on Speechmatics, please contact neilm@speechmatics.com

For more information on MiFID II, please see a recent Financial Times Article:


For a link to the IHS report: