Machine Learning Research for Customising Learning Content in Companies
Together with QuickSpeech, the St. Pölten UAS conducts research on the use of machine learning and natural language processing for the automated development and personalisation of learning content. This gives companies the opportunity to (further) train their staff in a playful manner. St. Pölten UAS graduate and QuickSpeech CEO Lukas Snizek developed the app during his studies and was guided and supported in his endeavour by the start-up promotion programme Creative Pre-Incubator.
High Demand for Further Education
As both the need for further training in companies and the demand for individualisation are growing, businesses are facing major challenges. Learning content has to be easily accessible and cater to individual needs. The start-up QuickSpeech enables easy access to internal company content and is now researching the automatic provision of personalised learning content together with the St. Pölten UAS. QuickSpeech is subsidised by the Austrian Research Promotion Agency (FFG) and focuses on machine learning in in-company learning.
Great Need for Further Training
“Fostering young talents and teaching entrepreneurial thinking are very important to us here at the St. Pölten UAS. Furthermore, we have been systematically supporting students in the development and implementation of their business ideas through the ‘7 Steps to Start-Up’ and the ‘Creative Pre-Incubator’ for several years now. We are therefore particularly delighted to see the newly created companies of our graduates develop so successfully – such as the up-and-coming company QuickSpeech. Our joint research and the further development of forward-looking ideas enable us to create added value for everyone involved, which shows that we are on the right track with our efforts for knowledge transfer and innovation”, says Hannes Raffaseder, member of the Executive Board of the St. Pölten UAS.
Personalised Learning Content
Under the auspices of Matthias Zeppelzauer, the cooperative project uses “Natural Language Processing“ (NLP) and Deep Learning for the automatic understanding of comprehensive collections of documents. On the basis of user interaction, the intelligent learning app understands the employees’ learning behaviours. At the same time, company training officers can upload source documents to the admin dashboard, the management tool of QuickSpeech. From these documents, the app automatically extracts suitable quiz questions and the corresponding answers.
“This allows for the independent generation of content in the QuickSpeech app. Quiz questions and answers no longer have to be created manually but can be derived directly from the data in a partially automated manner at least”, explains Zeppelzauer. One benefit of the research is that it takes work off the shoulders of human resources personnel who are responsible for compiling training content. Moreover, learning content is automatically created right where a need is detected based on the interactions – and it comes in the learners’ own usage patterns. For example, a given company’s apprentice may receive the content in the form of a mini podcast that is read out loud, while a store employee gets more visual material.
Research for Further Development
Lukas Snizek, who completed the bachelor degree programme Media Management in 2018, realised soon after the first presentations that his idea had potential. “The feedback was positive throughout. With the support of the St. Pölten UAS and the Creative Pre-Incubator, we are able to take the first important steps in the conception right away”, says Snizek. Since then, with the involvement of players such as PORR Group, Angelina Nathalie Karré and the Austrian Research Promotion Agency (FFG), the app has attracted important stakeholders.
“The mission of QuickSpeech is to make learning in companies not only more playful and interesting but also smarter. By systematically targeting gaps in occupational knowledge transfer, QuickSpeech improves not only the knowledge management but helps to develop individual potentials as well.”
QuickSpeech was founded in 2018. The company’s app represents an easy-to-use and intelligent learning platform that companies can use to further train their employees through messages, videos and playful quizzes. Businesses can either create their own learning content or use content created by QuickSpeech. Based on machine learning, the app adjusts to the needs of the individual employees and personalises the learning experience. With the support of several prominent customers, QuickSpeech won the Josef Umdasch Research Prize of the World Summit Awards in the category “Future Workplace”. In addition to a number of national and international prizes, the young start-up was also selected as the Austrian ambassador for the Social Innovation and Digital Learning Platform.