Text to speech (TTS) software, that is, software that converts the written word into spoken word via computer, is either a vivid example of man’s absolute laziness or a cool tool that’s fun to play with and has loads of potential for use in everyday life. You decide. Following are the facts.

Text to speech software converts the written text in Word documents, Web pages, PDF files, and E-mails into the spoken word. In other words, your computer reads written content to you by employing what they call a natural-voice character from the program. Several inexpensive text to speech programs are available via the Internet and one of the programs even converts text into audio, such as MP3 or WAV files, which you can then listen to on a CD player or iPod.

Makers of TTS software list various key advantages to its use.

• TTS software increases productivity, because users can perform other tasks, such as exercising or housework, while listening to the contents of their e-mails or other documents

• TTS software helps people avoid the eye strain associated with reading lengthy documents on a computer screen

• TTS applications can be used by authors for proofing their written correspondences, speeches, or literary works

• TTS programs benefit people seeking to learn a second language by giving them a chance to type in phrases and then hear the phrases spoken out loud

• TTS programs can help the vision impaired read content and better track words on their computer screen. Additionally, vision impaired people are aided by TTS instructions in bus stations, airport terminals, and other public places

• TTS software transmits real-time alerts for automated equipment, such as building alarms and public address announcements

Some programs include a variety of controls such as voice tempo, phoneme correction (for better, more realistic pauses and inflections on words), a variety of male and female voices, the ability to type in your own words and phrases, and the availability of multiple languages, such as French, German, Spanish, English, and others.

Who can benefit from using text to speech software? Think about the following scenarios: a lawyer in her office has her TTS software reading a long brief as she organizes files; an executive gets his annual report read to him while he multi-tasks at his desk; a teacher uses TTS software to proofread the week’s lessons while mopping her floor; a large airport uses a TTS system to automate all of its public service and flight announcements; a student, trying to learn French, uses his TTS software to watch the words and phrases go by on the screen as they are read out loud.

As for how the voices actually sound—well, they are much better than in the early days. In the late 90s, I worked for a software company in Santa Cruz California, Haptek, Inc., that used TTS technology widely in their programs and was a pioneer in the field. Haptek started out as a virtual reality arcade game company. Quickly, the founders of Haptek saw an opportunity to branch out into characters (now known as avatars) that could stream in real time over the Internet. Most of the voices back then sounded like computers—robotic and disjointed with an eerie, unnatural cadence to the speech patterns. Now, most of the programs I demoed for this article actually contained voices that sounded like real people, not machines.

In text to speech, an engine (which is a fancy way of saying a ton of computer programming), inputs the raw data—the text. Numbers or abbreviations in the text are converted to their full word equivalents, i.e., the software literally spells out numbers and abbreviations to form their complete word counterparts. This process is known as normalizing the data. Next, the engine develops phonetic transcriptions of the words and divides the words into their proper stress and intonation patterns or phonemes. Or, for the laymen, the software places the right emphasis on the correct syllables of each word to form normal-sounding clauses and sentences. Lastly, all the pieces are then synthesized and converted to sound. Or, as Wikipedia tells it: “The back-end—often referred to as the synthesizer—then converts the symbolic linguistic representation into sound”. The result is a computer that talks to you.

Text to speech software is an interesting way to convert written words into spoken ones and has lots of potential for use in everyday life. From automating public announcement messages in airports to uses in education and entertainment, text to speech software is not only useful, it’s fun. We’ll likely be seeing a lot more of it in the future. Imagine: your car telling you, in a soothing voice, that it needs an oil change. Yep, it’s coming.

###

Liz Casey of ButterFat Writing Services, Inc. (www.butterfatwriting.com) provides robust copy and technical writing for clients who want their written collateral to effectively communicate and make them money. She is a member of the Redwood Technology Consortium.

Copyright 2009, Eureka Times Standard Newspaper. The print version of this article first appeared in the 6/19/09 edition of the Times Standard.