Hello Felix!
Not sure what the license of the ASAP screen reader is? If I understand you correctly, it sends the text or an encoded version over a serial port to the host OS, in your case Windows? What is the license of the text to speech tool on the Windows size? Have people tried to do the same with Linux? There are free blind-friendly tools for Linux, such as Braille (BRLTTY) and text to speech output, Some distros make it easy to activate those. Once you have those, either use the general feature to make the contents of your DOS window audible (e.g. dosemu2 or dosbox). You can also run dosemu2 in terminal mode, so you can use the default settings for making the contents of any Linux terminal window audible. If I understand you correctly, you like the navigation and user interface of the DOS ASAP screen reader and would prefer to use that to decide which text should be sent to a separate text to speech in the host OS. That could be for example MBROLA, eSpeak, Festival, or FreeTTS. I remember having configured dosemu1 to capture printer port data while I was developing the "graphics" screen print tools for FreeDOS (ESC/P, HP PCL, PostScript) so I guess redirecting serial port data to some TTS would also be feasible with dosemu2. Whether ASAP can be included with FreeDOS depends on whether it is free and/or open source. Please tell. Yes, you can boot FreeDOS from USB flash drives on many computers which can boot from USB storage media. Whether you have write access to the USB stick from which you have booted depends on your BIOS: There have been reports of computers where no writes were supported or writes have been slow. Normally, I do expect it to work, with limited performance. Note that you must not change USB sticks after booting. If you use DOS USB drivers, you get more flexibility than when using BIOS USB boot drivers, but you will need a driver which supports your specific hardware. The biggest problem will be that you wanted a DOS text to speech system which can run as a driver and access HDA or AC97 sound hardware. This is highly unlikely to exist: Modern text to speech needs a lot of RAM and only a few apps such as MPXPLAY do have the complex sound drivers needed for HDA etc. So the text to speech would have to live in some protected mode ecosystem, making it DOS instead of the speech synthesizer being the background task. Which in turn would be similar to the more obvious option of using a Linux or Windows speech synthesizer and letting DOS run inside a window, which also avoids having to port a synth to DOS. Of course you could also work with 2 computers: One running plain DOS and the other, connected via serial or bidirectional parallel port, would run some text to speech system on any other OS. That could even be a Raspberry Pi, for example. Regards, Eric > Long story long: Joseph Norton put together a FreeDOS bootable > installation ISO with a DOS screen reader, becoming my hero in the > process, and using this I was able to install FreeDOS on a virtual > machine using a Windows-hosted speech synthesizer emulator listening > on a virtual serial port. ASAP, from deep within DOS, sends its output > to a serial port of my vm which is mapped to one end of a pair of > COM0COM ports. On the other end I have a speech synthesizer emulator > picking up that output and transforming it into actual speech using > ESpeak... _______________________________________________ Freedos-user mailing list Freedos-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-user
