The proliferation of mp3s and YouTube videos with poorly produced audio demonstrates to me that for most of the public, accessibility and convenience is more important than especially high quality sound. Do you agree? To give another example, the sound quality of most mobile phone calls I make is much worse than an old-fashioned landline. I never get halfway through a landline call and have to ask for a mobile phone number to ring back on! (but the converse case happens frequently). Some new results from the Calls of the Wild experiment supports my assertion about sound quality.
Last year I got involved in the birdsong experiment for the British Science Association. Mostly, this was about comparing our responses to pleasant warbling and ugly squawking to support the research of Eleanor Ratcliffe. But at the same time there was an experiment running for the Salford’s Good Recording Project which we didn’t talk about at the time, because otherwise the results would have been ruined.
We covertly added traffic and/or wind noise to some of the birdsong recordings to see how that might change people’s judgements. Here is some examples for a pleasant songbird:
With added noise:
People were asked questions about the birdsong (e.g. pleasantness), but none of these ratings were changed by the addition of noise. People focussed on the foreground sound and either didn’t notice the background noise or weren’t bothered about it.
We re-ran the experiment a few months later, but this time we added a question on the survey specifically asking about the quality of the recording. Once we drew people’s attention to audio quality, we observed a negative effect on bird song ratings for the highest level of wind noise (e.g. the noisiest example above).
Traditionally, audio quality research has focussed on listening for small degradations of signals using trained listeners. I think our results show that to the general public, sound quality is less important than audio engineers might hope. Am I wrong? please comment below.
We have another audio experiment running: Audio Battle. Please have a go, we need more subjects to complete the test. This time no tricks are being played, what it says in the instructions is exactly what is being tested.