Fluent.ai’s speech interface solution works in most languages and accents, allowing users to use their natural speech. Fluent.ai’s patented approach sidesteps the need to convert speech to text and instead extracts intent directly from speech.

Fluent.ai’s software is based on artificial intelligence technology that was developed by Canadian and Flemish universities. The CLiPS Research Group (Computational Linguistics & Psycholinguistics) of UAntwerp delivered parts of the technology and contributed to this international spin-off company.

Textgain was established in 2015 by dr. Guy De Pauw, dr. Tom De Smedt and prof. dr. Walter Daelemans and markets the technology developed at the CLiPS research center.

The amount of digital text that is produced nowadays is unrelenting: we are bombarded with newspaper articles, e-mails, whatsapp messages, tweets, and status updates. Never before have more pieces of information and opinions been published than today. And there is no-one that can contain the totality of the knowledge that is encoded in this massive amount of language and text. But with the help of automatic text analytics we can leave this task to the computer: it is this kind of technology that is developed by Textgain, a spin-off of the CLiPS research group (Department of Linguistics).

How does Textgain work

Textgain offers a lightning fast API that can perform linguistic analyses in real time: you can find out what language a text is written in, its text type or what the most important keywords of the message are. Textgain’s sentiment analysis for a wide range of languages additionally discovers whether someone is communicating positively or negatively about a specific product, service or concept. With such tools at your disposal, you instantaneously transform Twitter into a survey service that observes and measures opinions on a global scale.

And the technology can even read between the lines: personal writing style reveals a lot about a person’s demographic features. Textgain offers text analytic modules that can make an educated guess regarding an author’s gender, age, level of education and even personality type. Combined with sentiment analyses, this type of author profiling technology provides data that are invaluable to e-marketeers, digital publishers and data miners in general.

Custom-made text analytics

Textgain also develops custom-made text analytics solutions: in collaboration with Treecompany, Textgain monitored and analyzed the public debate on education on social media. They also developed a system that identifies jihadist propaganda on social media. In the years to come, Textgain will continue to market their technology and know-how in the domain of Big Data: as a spin-off of the Faculty of Arts, Textgain ties in perfectly with the other UAntwerp spin-offs in data mining at the Faculty of Economics (Predicube) and Computer Science (Froomle).

In 2001, prof. dr. Walter Daelemans (CliPS research centre) started the spin-off Textkernel, together with fellow researchers from the universities of Amsterdam and Tilburg. In their academic collaboration, they examined the way machine learning processes natural language production.

The job search has increasingly simplified with the digitising of job ads. This has caused a problem for well-known organisations: hundreds of candidates apply for the same function. Besides that, recruiters now have access to enormous databases to find the right employee for a company or vice versa. The HR-sector possesses so much information, out of which the need for computer automatising has risen.

Textkernel gives an answer to this need with its specialisation in semantic recruiting technology, which makes it possible for computers to transfer text into useable, structured data.

This language technology can for instance extract the most important information from hundreds of application letters and recognise the wanted experiences and skills. Textkernel offers solutions for the HR-sector to bring potential employers and employees together quickly.