Listening survey for product sounds

Description

A survey in which listeners audition candidate product sounds under controlled playback, rate each from zero to four stars and comment, giving a simple grade per sound.

People asked about creative work tend to be polite, so Case and Day propose a structured listening survey for product sounds, alerts and brand soundtracks. Testers audition every candidate, give zero to four stars and leave written comments. The crucial control is playback: online, instruct a set volume (for example 40% of device maximum) and specific earbuds, or record which headphones each person uses and analyse ratings by hardware; in person, issue identical headphones and devices and ask people not to change volume. Ratings are averaged into an overall grade per sound and presented in a short, self-explanatory summary, because such summaries travel without their presenter. It is a less formal relative of laboratory listening tests such as MUSHRA, trading psychoacoustic precision for speed and stakeholder persuasion.

Attributes

Loss type
Latent
Data
Mixed
Effort
Days

Steps

  1. Prepare the candidate sounds and a listening page or in-person station, with sounds in randomised order.
  2. Fix playback conditions: specify volume (for example 40% of device maximum) and headphone type, or record the hardware each tester uses.
  3. Have each tester audition every sound, rate it from zero to four stars and write a short comment.
  4. Average ratings into a grade per sound, segment by hardware if collected, and pull representative comments.
  5. Present a brief, self-explanatory summary so stakeholders can drop poorly rated sounds.

Sources

  • Designing with Sound (Case & Day, 2018)

Also known as

  • Standardised-playback listening survey
  • Online listening survey
  • Sound audition survey
  • Star-rated listening test

Categories

  • Analytics & Behaviour
  • Survey & Psychometrics