hi there was trying to see if i can compile julius speech recognition for openwrt.. i want to make a simple low cpu Ai interface for openwrt using julius.
compiling on linux one would run this
% sudo apt-get install build-essential zlib1g-dev libsdl2-dev libasound2-dev
% git clone https://github.com/julius-speech/julius.git
% cd julius
% ./configure --enable-words-int
% make -j4
% ls -l julius/julius
-rwxr-xr-x 1 ri lab 746056 May 26 13:01 julius/julius
I am stuck, I do not know what to do with these zlib1g-dev libsdl2-dev libasound2-dev and how to insert into the openwrt build make file or if i install outside the build on the host computer
basically if i can compile julius' then I just have a simple remote push button to enable julius listening for 10 seconds ( say my phase) and then let process it into a string output and then if it matches a predefined phrases do an action either via domoticz of an direct software operation on openwrt
any help would be appreciated as I still do not get how to work with openwrt makefiles very well and compiling software with it
You'll need to cross-compile your program, the (relatively) easiest way to accomplish that, would be to package up julius (and all its (missing-) dependencies).
thank you for the reply - not exactly sure what you are telling me. are say whatever missing dependencies from out of zlib1g-dev libsdl2-dev libasound2-dev and move them too a src directory in julius ..
question: how would I determine which file I would require ? as never really did that before either
okay so I tried compiling ran itno a coulple issues maybe some could help me with
when trying to compile I get this error
checking for grep that handles long lines and -e -- I checked grep it supports -e but I guess not long files
so I tried another route - i made docker container of julius server
and i am just trying to compile adintool but I get this error which i can not figure out
another issue is libgomp.spec no file or directory
adnitool are network interface tools that will connect the alsa mic or send files to converted if I could that work it be atleast a work around for docker container
as I can not seam to find libcomp.spec anywhere on the host
any help or suggestions would be welcome