PlainTalk is a collection of software that enables your Macintosh to speak written text and respond to spoken commands. PlainTalk 1.5 includes three components, each with its own installer: English Text-to-Speech, Mexican Spanish Text-to-Speech, and English Speech Recognition.
This document describes each of these three software packages, and discusses the differences between PlainTalk 1.5 and earlier versions of PlainTalk.
For more information about Apple Computer's speech software, please visit Apple's Speech web site at <http://www.speech.apple.com/>.
English Text-to-Speech
With the English text-to-speech software installed, many Macintosh programs can read English text out loud. To install the English Text-to-Speech software, you need the following software and hardware:
• any Macintosh or Power Macintosh computer
• version 6.0.7 or later of system software
• at least 300 kilobytes (K) of RAM in addition to what the rest of your system software requires
• at least 5 megabytes (MB) of space available on your hard disk
On System 7.0 or later (System 7.1 or later if you have a computer with less than a 68020 microprocessor) the Text-to-Speech software includes a Speech control panel. In PlainTalk 1.5 and later, this Speech control panel includes a "Talking Alerts" options panel. If you use this to turn the Talking Alerts feature on, your Macintosh will speak the contents of alerts to you. (An alert appears, for example, when your printer runs out of paper while printing.)
For more information about the English Text-to-Speech software, see the Using English Text-to-Speech document in the English Text-to-Speech folder.
Mexican Spanish Text-to-Speech
With the Mexican Spanish text-to-speech software installed, many Macintosh programs can read Spanish text out loud. To install the Mexican Spanish text-to-speech software, you need the following hardware and software:
• a Power Macintosh computer or a Macintosh computer with a 68020, 68030, or 68040 processor
• version 7.0 or later of system software
• at least 1.3 MB of RAM in addition to what the rest of your system software requires
• at least 2 MB of space available on your hard disk
For more information about the Mexican Spanish Text-to-Speech software, see the Using Mexican Spanish TTS document in the Mexican Spanish Text-to-Speech folder.
English Speech Recognition
With the English Speech Recognition package and the proper hardware, your computer can respond to spoken commands. The English Speech Recognition package includes the Speech Recognition system extension, the Speech control panel, and a utility called Speakable Items. With Speakable Items, whenever you say the name of an item inside the Speakable Items folder (which is inside the Apple Menu Items folder), the computer acts as if you double-clicked the item.
You can turn Speakable Items on and off using the Speech control panel. You can also use the Speech control panel to set several preferences for how Macintosh Speech Recognition programs listen and respond. For example, you can set the Listening options in the Speech control panel so that the recognizer only listens while you hold down a key of your choice. When you use this option recognition accuracy is improved and misfires (when the computer thinks you spoke a command but you didn’t) are eliminated.
To install and use the English Speech Recognition software, including the Speakable Items software, you need the following hardware and software:
• a Power Macintosh computer
• system software version 7.5 or later
• a 16-bit microphone, such as the Apple PlainTalk Microphone or the microphone built into some Apple audiovisual displays.
For more information about the English Speech Recognition software and the Speakable Items utility, see the Using Speech Recognition document in the English Speech Recognition folder. You can also learn more about Apple's speech software -- and check for updates -- by visiting Apple's Speech web site at <http://www.speech.apple.com/>.
Differences between PlainTalk 1.5 and PlainTalk 1.4.1
PlainTalk 1.5 is a relatively minor update to PlainTalk 1.4.1. It contains a couple new features and a few bug fixes:
• The Speech control panel in PlainTalk 1.5 includes a new feature called Talking Alerts. If you use the Talking Alerts options in the Speech control panel to turn the Talking Alerts feature on, your Macintosh will speak the contents of alerts to you. (An alert appears, for example, when your printer runs out of paper while printing.) The Speech control panel is installed when you run any of the PlainTalk 1.5 installers -- English Text-to-Speech, Mexican Spanish Text-to-Speech, or English Speech Recognition.
• The English Speech Recognition installer for PlainTalk 1.5 installs a version of the Speech Recognition extension (version 1.5.1) that allows Speech Recognition to run on all shipping Power Macintosh computers and PowerPC-based PowerBooks with 16-bit sound input. Version 1.4.1 of the English Speech Recognition installer installed an older version of the Speech Recognition extension (version 1.4.1) which did not work on some PowerPC-based Performas and PowerBooks that had shipped after PlainTalk 1.4.1 was released. Users of those computers had to separately install an updated Speech Recognition extension (version 1.5). That is no longer necessary.
• The Speakable Items utility (which gets installed when you run the English Speech Recognition installer) contains a few new default speakable items, including "Quit Application" and "Zoom Window".
• The Speakable Items utility now accepts numbers in item names. For example, if you make an alias to an application named "Super Application 4.3" and drag the alias to the Speakable Items folder, then you can open that application by saying "Open Super Application four point three", or "Open Super Application four three" or simply "Open Super Application". In earlier versions of the Speakable Items utility, numbers in item names were ignored.
• The "listen only while key(s) are pressed" feature (also known as the push-to-talk feature) in English Speech Recognition is now even more tolerant of noise or speech that occurs before the push-to-talk key is pressed, requiring little or no pause before pressing the key down and speaking a command. Using the push-to-talk feature increases recognition accuracy and eliminates misfires (when the computer misinterprets sounds not intended for the computer).
• The English Speech Recognition installer for PlainTalk 1.5 installs a version of the Speech Recognition extension (version 1.5.1) which has features and bug fixes that help developers create more robust speech recognition applications. Developers can learn how to use speech recognition in their applications by visiting Apple's Speech web site at <http://www.speech.apple.com/>.
Differences between PlainTalk 1.5 and PlainTalk 1.4
PlainTalk 1.4.1 was a minor update to PlainTalk 1.4. It included a few bug fixes, including the following:
• Version 1.4.1 and later of the Speech control panel (which is installed by version 1.4.1 and later of English Text-to-Speech, Mexican Spanish Text-to-Speech, and English Speech Recognition) contains a fix to incompatibilities that version 1.4 of the Speech control panel had with HyperCard 2.3's popup menus when running on PowerMacintosh computers.
• Version 1.4 of the Speech control panel failed to run on some 68000 Macintosh computers (including Macintosh SE computers). Version 1.4.1 and later of the Speech control panel runs on all Macintosh computers with system software version 7.1 or later.
Differences between PlainTalk 1.5 and PlainTalk 1.3
Version 1.3 and earlier versions of the PlainTalk software had a single installer program that installed both text-to-speech and speech recognition. PlainTalk 1.4 and later contain three separate installers to make it easier to install just the parts that are appropriate for your needs: English Text-to-Speech, Mexican Spanish Text-to-Speech, and English Speech Recognition.
You don’t have to install them all. However, if you are going to install version 1.5 of English Speech Recognition or Mexican Spanish Text-to-Speech then you should first install version 1.5 of English Text-to-Speech, even if you already have an older version of English Text-to-Speech installed.
English Speech Recognition version 1.5 is a relatively minor update from versions 1.4 and 1.4.1, as described above. But English Speech Recognition version 1.5 (like versions 1.4 and 1.4.1) has some significant differences from the speech recognition software in version 1.3 and earlier versions of PlainTalk. Some features were eliminated to simplify speech recognition and make it more robust and usable. Other options have been added to increase accuracy and eliminate misfires.
• Version 1.4 and later offers a new listening option in the Speech control panel, “Listen only while key(s) are pressed,” that lets you leave Speakable Items on all the time without worrying about misfires. It also improves the accuracy of recognition.
• Version 1.4 and later works in a wider range of environments thanks to its improved adaptation to room acoustics.
• Version 1.4 and later does not include or support Speech Macros, the Speech Macro Editor, or Speech Rules. It still supports speaking dialog buttons and the Speakable Items folder. Any item you place inside the folder becomes speakable—saying the name of the item launches or opens the item. You can use the AppleScript ScriptEditor to create scripts that are launched by speaking a command: just save the scripts inside the Speakable Items folder. To see examples, look inside the Speakable Items folder.
• Version 1.4 and later runs on Power Macintosh computers only, and does not support Macintosh AV computers with 68040 processors. You can continue to use PlainTalk 1.3 on 68040 Macintosh AV computers.
• Version 1.4 and later does not include automatic support for recognizing the spoken names of menu commands. You can still manipulate scriptable programs, such as the Finder, using AppleScript commands. To see examples, look inside the Speakable Items folder.
Deciding whether to upgrade to version 1.5 if you are currently running version 1.3 or earlier
In general, version 1.5 offers a more robust and flexible set of features than version 1.3 or earlier versions. However, some people in special circumstances may do better with version 1.3.
• If you have a Macintosh AV computer that has a 68040 processor and you use PlainTalk 1.3's speech recognition capabilities, Apple recommends that you continue using version 1.3 of the PlainTalk software. Though the new text-to-speech software works with these non-PowerPC-based computers, the new speech recognition software does not (because it runs on Power Macintosh computers only). It is not recommended that you mix versions, using the old recognition software with the new text-to-speech software.
• If you make extensive use of PlainTalk 1.3's Speech Macros, the Speech Macro Editor, or the ability to speak menu commands, then you might want to continue using version 1.3. As part of the effort to make the software more robust, version 1.4 and later of English Speech Recognition supports Apple's standard AppleScript ScriptEditor instead of the Speech Macro Editor, and it does not support the ability to speak menu items.
