From close listening to distant listening: building tools for speech-music discrimination in Danish P3 radio
Danish music radio has evolved dramatically since digitization — but how accurately can we measure the shift from music to talk? This study compares painstaking hand annotation of Go' Morgen P3 episodes with a Convolutional Neural Network applied to more than 65,000 hours of P3 broadcasts. Achieve blazing fast classification at 98% accuracy. Discover why the music-introduction tradition vanished, and what large-scale audio analysis really adds to media research.