Sound design with everyday words
Musicians often use non-technical words such as 'warm', 'sharp', or 'sweet' to describe sound quality. Commonplace experience indicates that the descriptions of a sound by a diverse group of musicians may vary, suggesting that musical background may influence one's interpretation of a sound. A research study was carried out targeting 844 subjects of varying musical backgrounds, where each subject had the chance to match various words to sound samples. Each subject was assigned to one or more musical categories (strings, woodwinds, electronic, percussion, brass) based on previous musical experience, and the results were compared across categories. Statistical measures were employed to determine if a correlation existed between musical background and survey responses. After analyzing the results from all sound-word combinations, it was determined that the musical background had no effect on the selection of words. Because of the nature of statistical hypothesis testing, the expected rate of false positives was greater than the proportion of statistically significant sound-word combinations. From the data collected in this user study, it is reasonable to suggest that the description of musical sounds is an innate skill that is not influenced by musical background and training.
In another test 40 sounds were presented randomly to each user from a list of 200 that were collected from the Freesound online database. After listening to each sound, users could enter free text as well as rate a list of words on a scale ranging from “not applicable” to “high”. 40 words were carefully selected from previous studies and from the labels assigned by contributors to the Freesound database. Out of those, 12 words were randomly presented through each survey session. Chosen words included material properties (e.g. wood, strings), adjectives from sensory modalities other than hearing (e.g. sharp, sweet), and subjective impressions (e.g. pleasant, noisy). Our preliminary study indicates that people tend to assign similar words to certain types of sounds, and thus suggests a fairly universal mapping between words and timbre. Moreover these words strongly correlate with audio features both in the frequency domain (i.e. spectral content) and in the time domain (i.e. amplitude envelope shape).
