The UK’s only text-to-speech (TTS) solutions company, CereProc, has launched version 3.0 of its core synthesis engine, CereVoice, updating the service to enhance accessibility and performance, while offering voices with character and personality in more languages to extend its reach in global markets across multiple platforms.
Text-to-speech specialist CereProc has unveiled version 3.0 of its core synthesis technology, introducing a range of features including iPhone compatibility, improved performance and support for an enhanced range of languages.
Launching version 3.0 of its core synthesis engine, CereVoice, the UK’s only text-to-speech (TTS) solutions company has updated the service to enhance accessibility and performance. In a release designed to extend the product’s reach in global markets across multiple platforms, voices with character and personality have been included in more languages than ever before, while the introduction of support for iOS brings the technology to mobile phones and tablets for the first time.
“We’re delighted to be able to introduce CereVoice to a whole new international audience with the introduction of three major languages. Our mission has always been to make this an international offering, and this release is an important step forward in our strategy to offer a truly global solution,” said CereProc CEO Paul Welham.
“This is an exciting time for the company, and we’re looking forward to talking with potential new partners in France and Italy who are interested in deploying or distributing high quality text-to-speech systems in their local markets.”
Responding to a blend of increased opportunity and global demand, CereProc have added French, Spanish and Italian to the existing language portfolio of English and German, while also increasing the number of English language accents to include US male and Scottish male voices. The updated diverse range of voices and accents serves to enhance user experience, optimising performance and reception of the realistic speech solutions among local or regional users.
The increased choice of languages heads up a range of new features and upgrades in CereVoice 3.0:
- Voices now support 64-bit versions of Windows, SAPI for .NET 4, and the latest version of OS X, Lion
- New CereVoice Software Development Kit (SDK) for Windows, OS X and Linux, with a simplified speech API, allowing developers to integrate CereVoice into their application with just a few lines of code
- New iPhone OS (iOS) SDK, bringing CereVoice to mobile phones and tablets for the first time
- Faster, more accurate text processing engine, with support for additional SSML and VoiceXML features
- Improved cServer performance allowing more channels and faster speech synthesis, even on commodity hardware (up to 340 x real time performance on Intel Core i7)
- New MRCP standard (Media Resource Control Protocol, v1 and v2) cServer Connector, simplifying deployment of our robust, high performance, multi-channel cServer on telephony platforms
- Improved speech compression for crisper, clearer audio quality up to 48khz
Although the new English, US and Scottish voices have been added to CereProc’s commercial library, the advanced speech technology experts have made the Scottish male voice available free to schools in Scotland via the Scottish Voice website at http://www.thescottishvoice.
Paul Welham added: “While the voices that bring their devices alive are the natural focus for our users, this release has also been very much about constantly improving the technology that makes it all run. These upgrades have been designed to make CereVoice an even more robust, more powerful tool.”
As the UK’s only text-to-speech (TTS) solutions company, CereProc produces speech synthesis systems that offer voices with character and personality. Having developed the world’s most advanced TTS technology, with consistent international commercial and academic success, the Edinburgh-based company is a recognised leader in all aspects of speech technology.
Founded in 2005 by a leading team of speech experts, CereProc works with a raft of companies and academic institutions to develop innovative TTS systems, designed to provide an engaging user experience, and delivering realistic, emotional, custom built voices in multiple languages.
Available on platforms from mobile and embedded devices to desktops and servers, CereProc’s core synthesis engine, CereVoice, remains at the forefront of TTS technology. Recently granted a patent for Emotional Speech Synthesis, the global TTS experts are relentlessly committed to innovation in research and development, continually pushing market boundaries to expand their product range and service areas.
CereProc offers its customers, from consumers and high profile individuals to world renowned institutions and brands, a wide range of solutions including multichannel TTS software for servers, application development tools, and custom built branded voices, alongside voice bundles for resale and individual voices for direct download.