Founded in 2012 as a spin-off of the Technical University of Munich, audEERING is a leading innovator in the field of AI-based audio analysis, specializing in voice analysis. Machine learning such as Deep Learning is used to develop neural network-based algorithms and train them using promising methods such as Unsupervised Learning. audEERING’s patented technology makes it possible to use a variety of different models for automated voice and general audio analysis. In this way, human characteristics and states, health-related voice biomarkers and sound events, and up to 21 acoustic scenes are robustly and reliably recognized in real-time.
The core product devAIce® provides as SDK, Web API, and plug-in – for Unity and Unreal – the ability to recognize speaker states and features in a variety of implementations and in full compliance with privacy regulations. The dimensional model used in this process analyzes activation, valence, and dominance, among others, to obtain results about the emotional impression of a speaker’s expression. In addition, age group and gender recognition are possible.
Through audEERING’s AI platform AI SoundLab, vocal biomarkers can be reliably analyzed in the context of studies, research and specific questions as well as longitudinal observations. It leads to improved insights in the field of consumer and patient satisfaction and wellbeing.
Find out more about audEERING on their website