Does anyone in here have experience working with Cepstral? I’d like to make my Cepstral voices sound more realistic. For example, voices at neospeech.com sound much more realistic. Can someone please help me?
Thats a generic sounding question. I have done a lot of work with Cepstral in all sorts of settings and have never had issues with the quality of sound … I have even had callers respond to Cepstral interactive sessions where the caller has mistaken a Cepstral voice for a live human on multiple occasions. Can you be more specific with what you are looking for or are you just a troll pushing an alternative product?
mudslide567:
Sorry I wasn’t very clear. I am creating .wav files using swift (swift -o test.wav “text, text, more text”) … then convert to .gsm with sox. I am using Asterisk to place outbound calls and Playback those .gsm’s … and the sound quality is decent but nowhere as good as Neospeech. Some words run together, you can’t understand other words, etc. So, it sound like you could possibly give me some tips?
Thanks - and no, not a troll 