Apple appears to have modified the audio of this week's WWDC 2026 keynote video whenever "Siri" was mentioned, apparently in an effort to prevent viewers' nearby devices from waking inadvertently during the presentation.
The technique was spotted by observers on X, who shared spectrogram screenshots showing clear gaps in those specific frequency ranges coinciding precisely with instances of the Siri name throughout the video. Apple appears to have cut out the 3kHz, 4kHz, 5kHz, and 6kHz frequency bands.
fun fact: tijdens de keynote hakt Apple een stukje 3k, 4k, 5k en 6kHz eruit wanneer ze "Siri" zeggen, zodat niet iedereens HomePods terug beginnen te praten 🗣️🚫 pic.twitter.com/x13WbNPztr — luuk de leest (@luuk58) June 8, 2026
The approach is designed to defeat wake-word detection, which relies on recognizing the acoustic profile of phrases like "Siri" and Hey Siri." By surgically removing the frequencies that carry key phonetic energy in the word "Siri," Apple can reduce the likelihood that HomePods, iPhones, iPads, and Macs in a viewer's home will trigger while the keynote plays back.
The technique does not appear to have been fully effective, however, as multiple viewers reported their devices activating anyway during the stream.
In 2017, Amazon was found to use a similar approach in its Alexa TV commercials, notching out frequencies to avoid triggering Echo smart speakers in viewers' homes.
