Assistive Speech and Language Analysis

The Assistive Speech and Language Analysis Group is working on the integration of speech therapy know-how into digital applications. At the centre of the group’s applied research are computer-assisted digital technologies for automatic language, speech and voice assessment. Here, the Oldenburg team is using and further developing its own technical methods for automatic speech recognition, acoustic voice analysis and digital signal processing as well as machine learning and computational linguistics.

In health research, the Assistive Speech and Language Analysis Group is making a significant contribution to counteracting the social withdrawal of persons affected by speech problems, for example due to illnesses such as Parkinson’s disease or after a stroke, and to helping them communicate. In addition, language features are applied as biomarkers, contributing in this way to the prevention and early detection of disease. The group’s language technologies and activities also support language education and literacy.

The team is also working on the accessibility for a broader public via digital learning platforms: Combining AI and mobile devices generates analysis and assessment tools for a wide range of health-related and learning applications, which are also available on a privacy-compliant server infrastructure.

Assistive Speech and Language Analysis in the field of Connected Health

Assistive speech and language analysis make a significant contribution to connected health and thus also to pioneering health research. To this end, Fraunhofer IDMT is working closely with the University Medicine of Oldenburg. In addition to voice and speech analyses, researchers are also studying and developing the recording of biosignals, e. g. with the SensOhr®-Platform, as a means to describe health and other vital parameters through body-worn sensors.

Findings from research and development are transposed into practice in the shape of mobile, sensor-based systems that can be used flexibly and without clinical infrastructure. As networked technological solutions, they facilitate improved diagnostics and therapy for specific diseases as well as digital language learning and support. For example, in the project »LeseKind« the Fraunhofer IDMT uses AI-supported applications for assessing reading proficiency.

Digital solutions for voice and speech analysis

Privacy warning

With the click on the play button an external video from www.youtube.com is loaded and started. Your data is possible transferred and stored to third party. Do not start the video if you disagree. Find more about the youtube privacy statement under the following link: https://policies.google.com/privacy

"Our vision is for everyone to be able to communicate with their voice," says Laura Tuschen, head of the Assistive Speech and Language Analysis group. The researchers are working on digital solutions to achieve this. In the video, Laura Tuschen explains how they are doing this and what needs to be taken into account, giving the OSST project as an example.

Insight into our reseach and practice

 

Press Release / 11.6.2024

Look who’s talking!

Intelligent technologies for analysing speech should detect anomalies even in nursery school children

 

News / 3.1.2024

Speech technologies that help us!

Laura Tuschen, head of Assistive Speech and Language Analysis group, gives an insight into the work of the research group.

 

Speech as a screening parameter

We develop a screening system for autism spectrum disorders an automatic analysis of speech as well as facial expressions.

 

AI-supported application for assessing reading proficiency

Use of automatic speech recognition technologies for assessing German reading proficiency among primary school students

 

Digital hearing and speech training

How can highly hearing-impaired people be supported in their therapy? We are developing a digital system to enable intensive training.

 

Automatic speech analysis

Based on machine learning, speech and voice pathologies should be able to be automatically evaluated and analysed.

 

Automatic Screening

With the »Oldenburg Speech and Voice Test«, we want to set a new standard for the automated detection of speech and voice disorders.

 

Intelligent speech training with ISi-Speech

In the ISi-Speech project, a training system is being developed that allows affected persons to independently train their speech capability. 

 

Connected Health

The work area Connected Health dedicates its research and development activities to sensor-based health systems that can be deployed flexibly and without clinical infrastructure.

Speech Recognition for Products and Processes

Speech detection systems allow operation of technical systems without visual or physical contact. The speech recognition technology of the Fraunhofer IDMT in Oldenburg functions robustly - with and without an internet connection.

R&D-Services and Licencing

Please contact us in case you are interested in our expertise and services.

All solutions at a glance

Here you can find further information about our solutions of the Oldenburg Branch for Hearing, Speech and Audio Technology HSA.