Asked
Resolved by Nomad 6R!
I'm working on a massively cool project at the moment. I'm hitting a wall as I source out the individual parts. I need one of the TTS voices to be post processed to sound more like a robot. There is one voice from Cereproc that is robot but I don't like the robot sound. I was looking for a windows program that would alter the voice as its spoken in realtime from EZ Bulider. Most of the programs I find are for altering a live mic, but they don't seem to effect the TTS voices. I've also looked into a post processing the voice on the speakers out, live with a toy placed inline after the computer to modulate the voice, but again i can't find anything anygood.
Any suggestions or good finds are welcome!?!
Related Hardware EZ-B v4
Related Control
Speech Synthesis
What I do is use the cool voices that Google has in the English download for Android phones or Google home speakers,There are many options to get the voice from natural down to many different robot sounds. Then I can use ARC mobile so that when a mobile script asks "say EZB wait" command for the default windows TTS it automatically converts to the robot voice on the Android phone that runs ARC mobile. I can then Bluetooth that voice to any speaker wired directly to a speaker on robot. I use the voice option Pink which is actually the lowest male voice from Google and then add the different values to get my robot sound now which is a close copy of the Starwars droids but lower to sound more cylon but very clear and understandable. It is only working with Mobile ARC though.
I am curious if there is a way to get this done...I wanted post processing for sound too, but never found a proper solution for Windows or for the Raspberry Pi!
Robo Rad...I'm impressed with your work around. I wont be able to use this set up for this particular robot.
Crazy that it's this difficult to find something to make this work. I've spent alot of hours on the internet. There really are not any good physical modulator machines that i can place after the voice to change it. Some really cheap kits from China is all I can find. Especially if looking into creating robot voices. There are alot of pitch voice changers out there (male to female and vise versa). I just want an effects box or software that can affect any input from the computer ie anything that comes thru the sound card.
hi will
i use swiftalker and cepstral.with swiftalker you can modified and make voices.
@nomad And these voices will work in ARC when you choose the voice with robot applied?
yes they work in ARC.you can make frases and download that to folder text.and in ARC you can choose from that folder. i wil make a video now so you can see and hear.
ok great. I just also found a program called Desk FX. https://www.nchsoftware.com/deskfx/index.html
It will alter any sound playing live. There are about 7 effects you can layer up to get a robot sound...BUT..if you want to play music or anything else NOT the robot voice...then those sounds will have the effects applied to them as well. Also not sure what the overhead will be on the CPU. I am using the panda.
This robot will have a finite amount of replies and spoken words. So maybe its just bringing 50 phrases into a sound editing program and have ARC play back the mp3 files. Just more work though.
EDIT: Now I can see thats is what nomad did...created mp3 from the program and triggers those in ARC. I was hoping for anything i typed in the speech recognition script would be spoken as the robot.
the link looks very pro,but is it working with ARC? cepstral has also a test page you can use to make a voice.you have to send them a mail asking for the link. it has also lots of settings.a few minites downloading the video. i am your father........will sounds great lol