YouTuber Geolocates Videos Using Airplane Sounds

YouTuber Colsto demonstrates how faint aircraft sounds in a video can help identify where it was recorded. As planes pass, the Doppler effect shifts the pitch of their sound, creating distinctive curves in an audio spectrogram. Those curves can reveal clues about an aircraft’s speed, distance, and movement. Because many aircraft broadcast their positions, speeds, and altitudes through ADS-B, public flight-tracking services provide records that can be compared with the sounds.

In Colsto’s example, the only initial clues were a recording date and that the footage came from somewhere in the continental United States; the image itself showed only trees. He found two aircraft signatures, separated by 920 seconds. Matching the sounds against flight data produced roughly 70,000 possible locations, which he reduced to about 15,000 by considering the timing and the absence of other expected aircraft sounds. He then modeled possible Doppler curves and compared their details with the recording, using changes in aircraft speed, altitude, and heading to pinpoint the likely flights and location. A Google Street View check led him to the exact security camera. The method is technically impressive but raises privacy concerns: it needs useful context to narrow the search, yet background audio can disclose a location that a video creator never meant to reveal.