Amateur Operators Make History By Launching Doom via Satellite Voice Command

Satellite operations usually rely on digital packets sent over radio frequencies, but one group of amateur radio operators decided to try a conversational approach. By speaking directly into a microphone, they verbally instructed an Earth-orbiting satellite to launch the classic 1993 video game Doom. This achievement marks what appears to be the first time a spoken instruction has been successfully sent to and executed by a spacecraft in orbit. The Oslo-based team targeted the European Space Agency’s OPS-SAT PRETTY, an experimental cubesat launched in 2023. Transmitting the simple phrase, "PRETTY PLAY DOOM," the satellite successfully detected the command, ran the game, and downlinked a postcard image documenting the moment.

Typically, spacecraft receive instructions from Earth through telecommanding, where operators use specialized software to generate digital commands. Ground stations then beam these instructions to the spacecraft using high-gain antennas as encoded radio signals, which the onboard computer decodes before execution. During the July 27 experiment, the amateur operators tested their system twice. The first trial involved transmitting a synthesized voice message on a continuous loop, while the second test featured an operator speaking directly into a microphone. In both instances, the cubesat captured the radio signal, isolated the voice from background noise, and transcribed it using an onboard speech-to-text AI model.

The experimental payload on the ESA cubesat features a processing pipeline built specifically to handle incoming audio and recognize spoken words. While booting up a decades-old video game from orbit serves as an entertaining proof of concept, the technology holds practical implications for future space missions. Researchers suggest that voice-command capabilities could eventually allow a network of satellites to monitor for global distress calls, picking up emergency signals and relaying them immediately. Furthermore, the underlying technology could be adapted to train onboard models to detect specific astronomical phenomena like gamma-ray bursts or colliding black holes in real time.