Text to Speech is a menu option in the Speech preference pane that allows Mac OS X users to hear the words that appear in a text file or web page.
Text to Speech dialog box
Instructions for generating synthetic audio using Text to Speech are summarized below:
1. Select a digital voice from the drop-down list in the Speech pane.
2. Open the application you want to use for this purpose.
3. Use the mouse to highlight the text you want to hear.
4. Navigate to the application program menu.
5. Choose Services > Speech > Start Speaking Text.
For example, if you want to hear the words that are on a Safari web page, move your mouse to the Safari menu, then navigate through the hierarchy via the Services submenu. While I can use the Apple software to listen to TextEdit, Mail, and Safari documents, it is not supported using either Firefox or Microsoft Office.
All but one of the voices installed by default by Mac OS X sound robotic and are easily distinguishable from a real human voice. The sole exception is the ALEX voice.
TextAloud from NextUp.com is a Windows program that is capable of generating high-quality speech using synthetic voices sold by third-party vendors.
TextAloud splash screen
For testing purposes, I installed TextAloud on my Mac using Windows XP running under Parallels Desktop 3. After opening the TextAloud application, I auditioned several voices using a sample script that I wrote.
TextAloud sample script
The TextAloud voice that I chose was Audrey, a British-accented voice sold by AT&T Natural Voices™. You can hear what she sounds like by CLICKING HERE.
One way to generate an audio file from a TextAloud document is built into the software program: simply click the Speak To File button located on the TextAloud menu bar. Another technique is to use Audio Hijack Pro to record the audio as it plays inside the Windows virtual machine.
Audio Hijack Pro recording screen
In my tests I discovered that Audio Hijack Pro is able to record audio that’s being played simultaneously by a Windows virtual machine that’s running under Parallels Desktop 3. I noticed, however, that I couldn’t capture audio using this same Mac program if I upgraded Parallels to version 4 or ran TextAloud for Windows under VMware Fusion.