My project is a personal assistant, something like Amazon Echo. I am using api.ai for the AI part and Google Speech API as a TTS engine (text-to-speech). My problem now is how to change the voice of the speech api to a more human voice (or maybe a funny voice).
I already found the voice that i need in acapela box but as you all know you have to pay for their Saas, is there anyway that i can found a TTS engine with multiple voices or producing my own voice and add it to Google speech for exemple ?
I found finally a solution !! Ivona has a great TTS engine with multiple voices and Java SDK. They give you a free access to their api as long as you don't surpass 50000 request, and i think that's more than enough.