[LAU] minimize clipping in pico2wave text-to-speach

Dear linux-audio-user list, I am using pico2wave on Debian 8 to synthesize voice from text. I encounter the problem that in the wav file which is produced, there is clipping, regularly producing harsh scratching sounds. I've downloaded the source code of the Debian packages "libttspico0" and "libttspicl-util" (containing pico2wave) but I am an absolutely lay person (I'm not even sure how to install the modified source as a Debian package, but this problem I'll address later on...). Can anyone of you tell me which line of code I have to edit to make the voice production lower (and by that minimizing the clipping)? (or maybe how to increase the maximum volume capability of the wav file as an alternative?) Any help is appreciated. Apart from the clipping the pico voices are quite agreeable to listen to, so I really want to get this done. Regards! Amie