ZipSpeak is a talking mini-flavor of Linux for blind and visually impaired people, based on version 7.0 of the ZipSlack flavor and version 0.08 of the Speakup screen reader. ZipSpeak is designed to be easily installed on an existing MS-DOS or Windows system, so that the user can start using a talking Linux system with a minimum of difficulty.

The standard version of ZipSpeak consists of a ZIP file which contains the main flavor; a set of Linux kernels which have Speakup built in, one for each supported speech synthesizer; a set of boot disk images for these kernels, along with an image that doesn't include Speakup; a utility for making boot disks from these images; and some documentation. While the standard version is usable with any supported synthesizer, there are also customized versions of ZipSpeak in which the appropriate kernel for a specific speech synthesizer is included in the ZIP file, and the boot disk for that synthesizer is the only included talking boot disk.

What is Needed to Use ZipSpeak
Users will need a PC with an Intel 80386 processor or better, with at least four megabytes of RAM, though more RAM is certainly recommended. (For more information about running ZipSpeak with four megabytes of RAM, see the fourmeg.README file in the ZipSpeak flavor.) ZipSpeak requires at least 100 megabytes of hard disk space, though more is recommended. Users will also need one of the following hardware speech synthesizers: Accent-PC, Accent-SA, DoubleTalk-PC, DoubleTalk-LT, Speakout, or any synthesizer that is compatible with one of these. Finally, users will need a system running MS-DOS or any version of Windows, though Windows NT with the NTFS file system will not work.

What is Not Needed to Use ZipSpeak
Users will not need to back up, repartition, or reformat their hard disks to run ZipSpeak; it will install on existing MS-DOS or Windows systems (except for Windows NT with NTFS). A user will also not need a sighted person to help with the installation or startup of the system; this can be done completely independently with speech feedback.