Sound is a powerful source of information.

By training algorithms to identify distinct sound signatures, sound can reveal what a person is doing, whether it's cooking, vacuuming or washing the dishes. And while it's valuable in some contexts, using sound to identify activities comes with privacy concerns, since microphones can reveal sensitive information.

To allow audio sensing without compromising privacy, researchers at Carnegie Mellon University developed an on-device filter, called Kirigami, that can detect and delete human speech segments collected by audio sensors before they're used for activity recognition.

A new study reveals how people worldwide turn to song lyrics—not just melodies—as a powerful tool for coping with grief, loss, and emotional distress. Researchers found a strong connection between lyrical themes and emotional goals like reducing loneliness or finding comfort. Unlike musical elements such as tempo or harmony, lyrics were consistently linked to listeners’ psychological needs. The study underscores the vital emotional role of words in music and suggests practical applications for mental health support and therapeutic settings.