AEYE uses SaaS model, by which I thought AEYE generates and stores in the cloud audio files for each webpage of the client’s website. When a user access an AEYE-powered website, it will connect to the cloud and plays back the audio file to the user. This, of course would require a lot of disk arrays to store the webpages and a lot of bandwidth to transmit the audio files. Additionally someone will have to trigger the audio file generation process each time a webpage is updated. This could cause a lot of potential problems.
On second read of AEYE into, I believe I was wrong on that. Now I believe, AEYE only stores its interpretation software in the cloud, possibly a long with some parameter/configuration files for specific clients. When a user access an AEYE-powered website and requests audio interpretation, the website will connect to AEYE cloud and make a call to the software, which in turn interprets the webpage real time and presents it to the end user. This would require less storage and less bandwidth. But AEYE may provide ways to store end user preferences, links and other configuration files for registered end users.
AEYE uses Amazon Web Services to host its cloud. This is a plus.