A method includes: prompting a user to recite a story associated with a first target emotion; recording the user reciting the story and recording a first timeseries of biosignal data via a set of sensors integrated into a wearable device worn by the user; accessing a first timeseries of emotion markers extracted from the voice recording; labeling the first timeseries of biosignal data according to the first timeseries of emotion markers; generating an emotion model linking biosignals to emotion markers for the user based on the first emotion-labeled timeseries of biosignal data; detecting a second instance of the first target emotion exhibited by the user based on a second timeseries of biosignal data and the emotion model; and notifying the user of the second instance of the first target emotion.