These are the sort of items I’d usually put in the This Week in Sound email newsletter (tinyletter.com/disquiet), but I’ve been super busy, too busy for a new issue, and so at a friend’s suggestion I am initially noting some here.
▰ The team behind the audio for the Halo video games share process in an advance peek at Halo Infinite, due out later this year. It’s packed with interesting details, such as how the relative proximity of gunfire wasn’t a sufficient filter in earlier games, leading to innovation this time around: “The new Halo Infinite audio system detects all gun sounds frame by frame, and prioritizes them in a threat order to decide output sound volume for each gun.” To a degree this is a matter of noise and confusion reduction, and of audio as an informative aspect of user interface. It also maps to hearing’s role in self-preservation. Likewise, ambient noises are adjusted in-game to make them more lifelike: “A variety of factors feed into this system, combining gameplay states, time of day, location tracking, timers, and more, all working together to bring the environment to life. This gives us the ability to create a dynamic mix of ambient sounds that remains compelling and immersive the entire time you’re playing.”
And you have to check out video of the destruction of an old piano done as part of the sound design effort:
▰ Molly Moser reports on how high-speed frame-by-frame records of hummingbirds flying helped sort out how they emit their trademark hum. The tiny birds yielded terabytes of data: “Higher harmonic content throughout the wing stroke, they explain in the paper, results in a ‘buzz,’ while equivalent first and second harmonic content makes hummingbirds ‘hum,’ and dominant first harmonic content results in the softer ‘whoosh’ of larger birds.”
▰ The digital civil rights group Access Now expresses concern that Spotify’s reported “mood-recognition features” could lead to “misgendering” and “discrimination,” reports Lilia Dergacheva. This is based on a patent “to detect emotion, age and gender using speech recognition algorithms.”
▰ People who require computers to speak for them deserve their own individualized voices that reflect “who they really are” — that’s the subject of a podcast episode from the Index Project featuring Rupal Patel, found of VocaliD. https://theindexproject.org/stories/podcast-restoring-lost-voices
▰ Grant Suneson of MSN Money’s 24/7 Wall Street lists the 25 jobs most likely to damage your hearing. Musicians rank at the bottom of the list. Higher up are surgeons, shoe and leather workers, gaming (aka gambling) service workers, and, at the top of the list, emergency medical technicians and paramedics.