What Ulf means is that this is code that's designed to output speech from the audio hardware of some server locked in a room in a datacenter somewhere and the server probably doesn't even have loudspeakers attached to it.
In order for a web application to be able to produce synthesized speech that can be heard by the web user, the generated audio needs to be piped to the web user's (client) computer, where the user's audio hardware can play it. Assuming that THEY have attached loudspeakers (or at least something like headphones).
Since web HTTP protocols don't support that, what you're going to need to do is basically the same thing you'd do if the web application was generating PDF's or graphics - create a temporary file with the data in it and attach it to a URL that the client would then request in order to get the audio data and play it.
There should be options for TTS output in wav, ogg, and/or mp3 formats (among others), so you need to read up on that part of the package.
Customer surveys are for companies who didn't pay proper attention to begin with.