Ideally, Text-to-Speech voices should read texts out loud sounding just like humans. Unfortunately, despite recent improvements, most text-to-speech voices still have an artificial/synthetic sound and this lack of expressiveness makes them less appealing.
IVONA Voices allow lifelike, expressive reading of words, sentences, paragraphs and even entire publications. BrightVoice technology uses artificial intelligence algorithms, which reflect the expression and individual characteristics of the human voice. As a result, the synthesized speech is virtually indistinguishable from the professional voice-over recording. BrightVoice is a result of 10 years of research.
IVONA Text-to-Speech takes advantage of many other technological innovations that make it a revolutionary system in the TTS industry. Since 2006, IVONA has maintained a leading position in the prestigious annual scientific event Blizzard Challenge (comparison of Text-to-Speech systems).
RVD technology (Rapid Voice Development) makes the process of building IVONA Voices fast and relatively inexpensive. It uses a set of tools modeling linguistic characteristics such as subvocalization, accentuation and intonation. It also allows to efficiently, quickly and accurately detect the speech signal in original speech recordings.
The first voices created with BrightVoice technology are British English voices: Amy, Emma and Brian and American English voices: Kendra, Kimberly and Joey.
The British English voices were created in partnership with Royal National Institute of Blind People (RNIB), UK. Steve Tyler, Head of Innovation at RNIB says, "IVONA Software has delivered amazing quality and innovation in an area that we are particularly keen to support. We are proud to have met the team at IVO, who have worked extremely hard to help us achieve our goals. Their focus on quality shows in every aspect of our partnership.".
The American English voices were created in partnership with one of the most well-known recording studios GM Voices, USA.