Hi Peter, there already exists a QEMU command line option -nographic which disables graphical output and redirects the first serial port of the emulated machine to the console.
With QEMU running Linux in text mode, this works fine - you will "see" the boot messages and can run a shell or any text application from the console, so it should be possible to analyse this output with a screen reader. QEMU can also redirect any serial port to pipes, so some speech synthesizer like "festival" could be connected directly. This URL describes the command line options -serial and -nographic: http://fabrice.bellard.free.fr/qemu/qemu-doc.html#TOC10 I have no idea how you can access graphical interfaces like MS Windows. Stefan Peter Laursen schrieb: > Hi everyone, > > I have downloaded the 0.9.0 version of the Qemu source code and managed > to compile it using MinGW and MSYS. But I face a problem with the > compiled binary. > > I am blind and therefore need to redirect the output from the guest OS > to a standard Windows console or a software speech synthesizer so that > a screen reader will be able to read the output produced. What > functions will I eventually need to modify? A text console would of > course be preferred. > My main reason for hoping to getting this to work is to allow blind > people to experiment safely with other operating systems - it would > also minimize the learning curve. > > I hope that you will be able to point me in the right directions as to > which source files and functions I would need to change. I have > searched through the source files and googled for documentation for the > source code, but I didn't find any. I also searched to see if there > were any patches that might accomplish the work - I was unable to find > any. > > I look forward to hearing from you. > > Peter.