The term generative audio modelling belongs to the categories of artificial intelligence, digital transformation and digital society. This technology describes computerised systems that can independently generate audio content such as music, speech or sounds. This is made possible by so-called artificial neural networks, which are trained with large quantities of sound samples.
Instead of just playing back existing sounds, the AI system "thinks" creatively and creates completely new sounds or soundtracks. A typical example: you want a customised jingle for a podcast. In the past, you had to commission a musician to do this. Today, a generative audio model can create a suitable audio track directly based on your desired mood - cheerful, calm, digital - in just a few seconds.
Such models are used in music production, marketing and voice assistance systems. They save time, enable innovative ideas and are often cheaper than conventional production. The use of generative audio models fundamentally changes how we deal with music and sound - and makes high-quality audio content accessible to everyone at any time.
Deutsch 
English (UK)