speech synthesis (text to speech) help
Hi, party people! I run a radio show and I am sketching out an idea for a max patch that acts as an automated software DJ! I will attempt to teach it all of the music locations on my computer and not to play a song twice. It will run a program where it plays novel music (perhaps even using en_analyzer~ to determine which songs go together, but that’s a complex feature that i won’t consider now), play commercial breaks from file on time, and in between songs I want it to read text strings that I program into it about this or that. (maybe even tell it what to say using op.recognize!) The difficulty I’m running up against right now is finding a way for a max patch to read text aloud on a Windows computer. I would also like to apply audio effects to the voice, but again, this is a trivial secondary matter. The main issue is that i am finding all sorts of good text-to-speech patches for Macs, i would like to run this program on a windows laptop. Does anyone have any objects or externals lying around that work fairly well? a speak-and-spell type of voice would be really cool, but i would settle for the program just sending commands to Microsoft Sam. Any advice would be helpful. Cheers.
Aka.speech is really super but do not work on windows.
So I am looking for a simple equivalent text to speech object in max msp under windows, in english or even better in french, to use in applications for disabled.
If you have any new info on this topic, please let me know.
I am also very interested in speech synthesis for the aid of disabled. I know its been nearly a year since anybody has posted, but there are any new developments on windows based max/msp speech synthesis lets get talking : )
All the best!
there are many things on linux, mac, but windows i’m not sure tbh… maybe this works for Windows : http://mage.numediart.org/ if you are knowing anything about compiling form sources. I’m actually studying it at Numediart :) but we are really not trying to have it work on Windows, actually it would maybe be interesting if that were to be the case. I’ll see…
The problem with aka.speech is it seems to play out of the computer’s audio, not Max/MSP. Thus, I can’t edit the voice through MSP (for example, to give it a bit of reverb), nor can I play multiple voices at once. Does anyone know a way around this?
Kirkpearson : either use another text-to-speech solution, or use a sound rerouting utility like soundflower or jack
just tried aka.speech and aka.shell along with Orcutt’s shell. Max 7 does not like any of them. Is there a Max 7 version of the aka objects and/or shell?
@kirkpearson : actually, if you go in command line and type "man say" there is an option (-n) to send the output through "AUNetSend" ; which is an audiounit plugin that should be installed by default ; there is also an option for specifying the audio output port (-a, you can use that to route output to soundflower at least)
Forums > MaxMSP