What you told us about recording audio: an overview of our web survey.

In an earlier blog post we presented some findings from our web survey on the differences between iPhones and other brands of mobile phone. In this post we look beyond mobiles and give a brief overview of some of the other findings from the survey. If you’ve already taken part in the survey, thank you very much for your time and answers. If you haven’t (and you still can, by clicking here), we asked 10 multiple-choice questions about people’s everyday experiences of making audio recordings. These questions covered subjects such as what recording devices people commonly use, who/what the recordings are for, what the recordings are of, what people do with recordings once they’ve made them, what problems people encounter when making recordings (in terms of impaired quality), and so on.  Here is a brief summary of some results so far:

  • The most popular device for recording audio was the mobile phone. Second and third most used devices were specialist audio recorders and PCs/Laptops.
  • iPhone users were found to behave very differently from non-iPhone users (as far as recording audio is concerned); they record different material, in different places, experience different recording errors, and upload their recorded material to the web far more often.
  • Almost universally (across all devices/levels of expertise, etc) the most common problems affecting audio quality were:
    • 1) Background Noise (selected by 59% of sample)
    • 2) Wind-Induced Noise (47%)
    • 3) Handling Noise (32%)
  • The most common subject area for recordings to be made in was the category of Music Events (e.g. Concerts, Festivals, etc).
  • Respondents to our survey were predominantly making recordings for their own benefit/use/pleasure (as opposed to for friends, the general public, for work, etc).
  • This said, most people (nearly two-thirds) do upload at least some of their content to the web. Facebook and YouTube are the most common destinations. Of the websites without visual accompaniment to the audio SoundCloud was most popular.
  • By proportion of responses, specialist audio recorders were the device(s) most frequently associated with each and every type of audio degradation we listed. However, these were also the devices most likely to be used by experts/professionals, and, furthermore, the most likely to be used with headphones while recording.
  • One possible recording ‘error’ that respondents could select was that “speech is too quiet”. This option was disproportionately selected by those who also stated that they had “little or no knowledge” of audio recording.

 

So how do the survey findings help the project?

The Good Recording Project has a particular focus on audio quality in user-generated content. The survey helps us in that we can begin to develop a greater understanding of what, how, why and where people of different levels of expertise are making recordings – and what they do with them.

We have already launched a web experiment to investigate how audio quality is affected when speech is played in noisy conditions (you can – and should! – take part in it by clicking here). The survey provides reassurance that we started out in the right direction by choosing to focus on factors such as background noise, wind noise and handling noise as common impairments to audio quality.

The survey also hints at some interesting, slightly more speculative, conclusions worthy of future consideration. The finding, for instance, that (proportionally) experts report more recording problems is somewhat counter-intuitive. Perhaps it suggests that experts assess quality differently to non-experts in that they have the knowledge and motivation to notice and identify different sources of quality impairment. Perhaps their threshold for what constitutes ‘quality’ is simply higher. It is also noteworthy however that non-experts were most likely to report that “speech is too quiet” as a significant factor impairing quality in their recordings.  The same group of people were found to be the most likely to record in domestic scenarios – settings in which speech is likely to be particularly important. Perhaps non-experts’ assessments of quality are more closely related to speech and/or ‘function’ than experts’ are? These sorts of findings are important to help guide our future research in the right direction and to properly explore the differences in the ways experts and non-experts form quality judgements.