Why aren't dynamic phone ringtones more mainstream?

Question

Considering that we are being yelled at by electronic devices at a high rate, why don't we have dynamic, synthesized notification sounds in our electronic devices?

Our brains can pick up nuanced information from audio signals and we do it all the time, for navigation, communication and pleasure.

By means of good, generative sound design, it would technically be relatively trivial to use software synthesis to produce rich ringtone (and other signal) audio on the fly, which could convey meaningful information to a recipient.

I'm wondering why we don't see this more often?

Compared to visual signs, the way sounds are being used for symbolic communication seems somewhat primitive.

An average app/system has a shallow set of static sounds which lets you at best discern a handful of event types.

Sure, static rendered audio files are easier to distribute across platforms and are less taxing on the device resources. But modern devices have more than enough processing power for some intermittent near-realtime audio processing.

And of course, extracting information from sounds can fail in noisy environments, but visual fallbacks can help and it would be the job of a sound designer/programmer to produce resilient tones.

However, I believe that generative, contextual ringtones could be quite useful in a lot of ways.

Can you point out good examples of where such on-the-fly sounds are being used?

Why isn't this more mainstream?

Answer 1

Sonification is the general term for this, and has been a topic in HCI research for a long time. Your question brings up a lot of interrelated, complex issues, but in my opinion the basic question of "why don't you see this more" boils down to a combination of:

• a general lack of UI designer familiarity/experience/training with sonification as a technique
• a lack of robust tools to support the technique
• limited audio capabilities of mobile devices (one, tiny speaker means only tone/pitch/volume can be varied - no spatial audio)
• relatively few use cases that entail users 1) not being able to look at a screen and 2) being able to extract information out of an "audio icon" better than the spoken word.

The paper linked above has a much more detailed discussion of these and other issues, which I'd recommend as a better answer.

As for where you do see this (outside of research systems?), the Geiger counter is probably the canonical example, and I would argue cars piping in engine noise into the cabin would fall into this category as well.