Voice-output devices convert digital data into speech-like sounds. These devices are no longer very unusual. You hear such forms of voice output on telephones (“Please hag up and dial your call again”) in soft-drink machines, in cars, in toys and games and recently in vehicle navigation devices. Two types of voice-output technology exist: speech coding and speech synthesis.
Speech coding uses actual human voices speaking words to provide a digital database of words that can be output as voice sounds. That is, words are codified and stored in digital form. Later they may be retrieved and translated into voices as needed. The drawback of this method is that the output is limited to whatever words were previously entered into the computer system. However, the voice output message does sound more convincingly like real human speech.
Speech synthesis uses a set of 40 basic speech sounds (called phonemes) to electronically create any words. No human voices are used to make up a database of words; instead, the computer converts stored text into voices. For example, with one Apple Macintosh program, you can type in Wiyl biy ray5t bae5k-the numbers elongate the sounds. The computer will then speak the synthesized words, “We’ll be right back.” Such voice messages are usually understandable, though they don’t sound exactly human.
Some uses of speech output are simply frivolous or amusing. You can replace your computer start-up beep with the sound of James Brown screaming “I feel good!” Or you can attach a voice annotation to a spreadsheet to say “I know this looks high, Bob, but trust me”
But some users are quite serious. For the disabled, for example, computers help to level the playing field. A 39-year-old woman with cerebral palsy had little physical dexterity and was unable to talk. By pressing keys on the laptop computer bolted to her wheelchair, she was able to construct the following voice-synthesized message: “ I can do checkbooks first time in my life.”