Creating a more accessible audio CAPTCHA

[mbgower]

As discussed in #3198, where CAPTCHA is used, it would be good to generate guidance on how to create an audio alternative that better equates to the allowed object recognition for 3.3.7 Accessible Authentication.

One example given was an audio segment that has the sound of a bee, which the user must identify. Such an example would place less cognitive burden on a user, while still allowing a user who could not see an image-based recognition to pass the CATPCHA.

[kaveinthran]

It's a good suggestion,

how to make sure that the sounds are not to obscure, or unfamiliar, accounting cultural and the state of knowing for different people.