On 18/05/10 05:05, Platonides wrote:
Tim Starling wrote:
Audio CAPTCHAs, like visual CAPTCHAs, are not accessible for all people and do not conform to W3C accessibility guidelines. What's more, they're easier to crack than visual CAPTCHAs due to their one-dimensional nature. This is especially true if you use a public source dictionary of spoken phrases, against which an FFT correlation can be run.
Just as with image captchas, you'd need to introduce noise into it.
I have been trying flite, and didn't find the synthesized text too understable by itself. :(
You have to introduce enough noise into it to defeat the computer, but not so much as to defeat the human. I've done some experiments myself, and I've read some articles on audio CAPTCHA design, and I'm not convinced it's possible.
I've seen an open-source audio CAPTCHA that uses speech synthesis, it seemed to be just designed as a deterrent. Like MathCaptcha, it'll only work until someone could be bothered to crack it.
If you use a public dictionary of spoken phrases, then the task is correlation, which is probably easier for computers than for humans. If you use a secret dictionary, then the task is speech recognition, which is more difficult. But if the dictionary is too small, then it's vulnerable to reduction to correlation, using precomputed or human-solved phrases.
-- Tim Starling