University of SzegedOpportunity Contact
Enikő Koppány, Business Development Manager
Szeged, Hungary

University of Szeged Website
TherapeuticEarly Stage

The recognition of Mild Cognitive Impairment (MCI) is one of the most difficult tasks of clinical psychiatry. Current tests are very time and labor-consuming and can be applied only by experts. Speech Gap Test is able to help its users to determine the chance of dementia by applying automatic speech recognition with a fast and simple method.

Description of Technology
The recognition of Mild Cognitive Impairment (MCI) is one of the most difficult tasks of clinical psychiatry and is usually regarded as a prodromal stage of Alzheimer’s disease, which mental disorder may take years to develop and is really difficult to diagnose. The existing diagnostic test procedures (MMSE, Clock-Drawing, ADAS-Cog) are not able to accurately recognize the subclinical phase of the MCI. The main problem in the diagnosis of MCI is that the current tests are very time and labor consuming, complicated, furthermore they can be applied only by experts and cannot be used as a screening test. This problem not only occurs at pharmaceutical companies, but it is also an extremely difficult task for the health insurance companies when they screen the health status of the potential customer. However recent studies reported that MCI causes slight changes in the speech of the patient.

The technology called Speech Gap Test is a mobile application developed by the researchers of the University of Szeged, which is able to help its users independent of their language and education to determine the chance of dementia with around 80-90 % accuracy by applying novel automatic speech recognition (ASR) using the latest deep learning technologies. It can inform its user or the pharmaceutical/insurance companies about the risk factor of dementia and recommend visiting a specialist in order to get a proper diagnosis or it can automatically transfer the processed data to a doctor/healthcare institution.The determination of the chance of dementia is based on the measurement and analysis of acoustic parameters of spontaneous speech, such as duration of speech, number, and length of silent and filled pause (pauses where the speech sample is not silent but where speech is also not produced. These pauses, which are typically filled with some speech sound or filler, reflect the hesitation of the speaker/user. For example, filled pauses can be verbal hesitations such as ‘um’, ‘uh’, or ‘er’), the ratio of silent and filled pauses and the duration of the speech sample, etc. The silent pauses, filled pauses, and the aggregate number of pauses, as well as the combined duration of the pauses, shows a significant difference between the group with neurocognitive impairment and the control group.


  • works independently from the native language, education, and status
  • screening based on spontaneous speech
  • rapid, accurate, cheap, simple, objective, non-invasive, without side effects

Type of Business Relationship Sought
Out license, Partnership

Patent Information
PCT Filed

The pace of change in our world continues to accelerate.  We all see this play out on the grand stage daily:  new diseases, warming oceans, increased hurricanes and fires, supply shortages, changes in social behaviors, more focus on alternative energies; the list goes on and on.  
We all have a role in proactively preparing for the ongoing shifts that we will encounter.  Our passionate team at Cognis Group is focused on helping you navigate innovation potential for the greatest collective benefits to people, the planet, and long term profits.

For more information, contact us:
US: +1 303 709 3157
UK: +44 1904 435190