This is an experiment to raise questions about the potential impacts of aural personal media curation.

During the current lockdown period, the government’s main concern is the physical protection of all citizens. However, as the period extends, an increasing psychological burden appears due to media over-confrontation with Covid-19.

As a frequent radio listener, I was trying to find a solution to avoid the topic without cutting myself off. Therefore I added a curation layer on the consumer’s side: a personal audio content curator.

It’s a real-time pipeline which checks an online radio stream of choice via speech recognition for predefined words. It then replaces those with another custom sound/word. An analog FM radio signal is then transmitted throughout my apartment, thus overwriting my kitchen radio with a cleaned, self-protecting audio stream.

This way I can go back to listening to early morning radio, cleaned from stress-inducing influences.

How does it work exactly?

Illustration showing the infrastructure.

Digital Radio Stream → Speech recognition reacting to individual terms → Overwriting of unwanted terms with custom sound → Overwriting of local radio waves with an FM transmitter





One week short-term project at UdK


May 2020


C/O Berlin, DE, 2020