Timothy's summary of the article refers to Speakup as a "Linux speech synthesizer". Actually, it's a screen reader. A screen reader looks at the contents of the screen and sends them as text to the speech synthesizer. And Speakup only supports hardware speech synthesizers at the moment, so you can't use Speakup (and therefore ZipSpeak) with just a sound card.
The only reason why there is an "approved list" of speech synthesizers is that the people working on Speakup haven't written drivers for all known hardware synthesizers yet. If your company makes a speech synthesizer that isn't yet supported, you could make it supported by writing a driver for Speakup, and I will put out new releases of ZipSpeak as Speakup is updated. Neither I nor the original creators of Speakup meant to lock people into an inferior synthesizer. By the way, if you happened to be referring to the DECtalk speech synthesizers, the author of Speakup is working on a DECtalk Express driver, and I'm sure that will be in the next release of Speakup.
But you don't hear speech on the built-in PC speaker under Linux. If you use Speakup, then you use a separate speech synthesizer device, which is either a card that you install in the computer or a device that you attach to the serial or parallel port. The speech synthesizer has its own speaker. If you use Emacspeak, then software synthesizers are also available (MBROLA, IBM ViaVoice Outloud, and probably Festival sometime soon).
Actually, it is no longer true that everything can be done through a text interface under Linux, especially as more proprietary applicatiosn appear. Screen reading technology for the GUI under Windows has gotten pretty good, and I know some blind people use Windows very productively. We don't have anything like that under X, and the fact that X uses multilple GUI toolkits doesn't help. This is something I'm concerned about, and I'd like to work on a solution, but I don't yet know enough about programming under X to do it.
I also have a friend who uses Emacspeak and loves it, but as someone else has mentioned, there's a lot of learning involved. For someone who's already familiar with an MS-DOS screen reader, Speakup is a lot more familiar than Emacspeak. I know some people are coming over to Linux from DOS, and they want to run PINE, Lynx, and whatever else they used on their shell account at their ISP. While Emacs has a terminal emulator, Emacspeak has poor support for it in my opinion, so if you're going to use Emacspeak exclusively, it's much better to learn to use VM (the mail reader), W3 (the Web browser), and so on under Emacs. So for people who are laready comfortable with a DOS screen reader and a Unix shell account, Speakup would be the best option, in my opinion. Also, you can't run Emacspeak on a Linux boot/root disk, so unless a blind person has a screen reader like Speakup, they need someone who can see to help them with the installation. The same thing applies if something goes wrong in the boot process and they can't get to a login prompt.
Slashdot Mirror
Timothy's summary of the article refers to Speakup as a "Linux speech synthesizer". Actually, it's a screen reader. A screen reader looks at the contents of the screen and sends them as text to the speech synthesizer. And Speakup only supports hardware speech synthesizers at the moment, so you can't use Speakup (and therefore ZipSpeak) with just a sound card.
The only reason why there is an "approved list" of speech synthesizers is that the people working on Speakup haven't written drivers for all known hardware synthesizers yet. If your company makes a speech synthesizer that isn't yet supported, you could make it supported by writing a driver for Speakup, and I will put out new releases of ZipSpeak as Speakup is updated. Neither I nor the original creators of Speakup meant to lock people into an inferior synthesizer. By the way, if you happened to be referring to the DECtalk speech synthesizers, the author of Speakup is working on a DECtalk Express driver, and I'm sure that will be in the next release of Speakup.
But you don't hear speech on the built-in PC speaker under Linux. If you use Speakup, then you use a separate speech synthesizer device, which is either a card that you install in the computer or a device that you attach to the serial or parallel port. The speech synthesizer has its own speaker. If you use Emacspeak, then software synthesizers are also available (MBROLA, IBM ViaVoice Outloud, and probably Festival sometime soon).
Most, if not all, speech synthesizers will let you plug in headphones and listen privately, so that nobody else has to hear your computer talk.
Actually, it is no longer true that everything can be done through a text interface under Linux, especially as more proprietary applicatiosn appear. Screen reading technology for the GUI under Windows has gotten pretty good, and I know some blind people use Windows very productively. We don't have anything like that under X, and the fact that X uses multilple GUI toolkits doesn't help. This is something I'm concerned about, and I'd like to work on a solution, but I don't yet know enough about programming under X to do it.
I also have a friend who uses Emacspeak and loves it, but as someone else has mentioned, there's a lot of learning involved. For someone who's already familiar with an MS-DOS screen reader, Speakup is a lot more familiar than Emacspeak. I know some people are coming over to Linux from DOS, and they want to run PINE, Lynx, and whatever else they used on their shell account at their ISP. While Emacs has a terminal emulator, Emacspeak has poor support for it in my opinion, so if you're going to use Emacspeak exclusively, it's much better to learn to use VM (the mail reader), W3 (the Web browser), and so on under Emacs. So for people who are laready comfortable with a DOS screen reader and a Unix shell account, Speakup would be the best option, in my opinion. Also, you can't run Emacspeak on a Linux boot/root disk, so unless a blind person has a screen reader like Speakup, they need someone who can see to help them with the installation. The same thing applies if something goes wrong in the boot process and they can't get to a login prompt.