Voice Recognition

Software for the Future



Voice recognition software is quickly becoming the computer industry’s newest trend. Consumers are intrigued by the idea of just speaking to a computer to compose a letter, send an email, or even search the web. Microphones will eventually replace keyboards as the main input source for a computer. But before we jump into the future of this new technology, let’s examine its past.


*History*

The first recorded success with speech recognition technology was in 1922. A little toy dog named "Rex" made technological history by coming to his master’s feet when he heard his name. The pup was held in his doghouse by an electromagnet against the force supplied by a spring. The magnet would release the dog’s iron base when 500cps of acoustic energy was received and Rex would ‘spring’ from his home to the feet of his loving owner. Although the toy was a simple battery-operated mechanism, voice recognition was forever changed. The dog would only respond to the name "Rex", which holds the precise acoustic currents needed to deploy the magnet.
Radio Rex with his electromagnetic doghouse

Although Rex wasn’t the ‘Tickle-Me Elmo’ of the 1920’s, he didn’t go unnoticed. During the 1940’s and 1950’s, the United States government quickly took advantage of the voice recognition technology. The Department of Defense applied it to Russian radio transmissions for immediate interception and translation of signals.

After peace was restored, universities became the primary contributors to the voice recognition industry. Research and development was slow until the creation of "Speak and Spell", an educational toy produced by Texas Instruments in 1978. The toy contained a speech chip using technology called ‘digital speech synthesis’ that enabled it to speak certain words when prompted. The public acknowledged Speak and Spell, along with the entire voice recognition industry, after a cameo appearance in E.T.: The Extra-Terrestrial , when E.T. himself used the toy to ‘phone home’.

Dragon Systems, the first company devoted entirely to speech recognition development, was founded in 1982 by Dr. Jim Baker and his wife. Throughout the decade the company concentrated mainly on research with one ultimate goal in mind: to develop a program that could be used by everyday people on personal computers. Finally, in 1997, "Naturally Speaking" was introduced. It was the first continuous speech software available with such a large vocabulary. Not too long after Dragon Systems’ introductory program, IBM released VoiceType and then ViaVoice. The market was becoming familiar with voice recognition software, however the programs were not very easy to use and incredibly time consuming.


*How It Works*

Voice recognition software is loaded onto your computer the same as any other program. A microphone is then attached to the computer to be used as the main input device. The rest can easily be summed up by the following steps:

  1. The microphone picks up any audio signals the user sends its way.
  2. The computer’s sound card then receives the information. The voice recognition software can now analyze the audio and distinguish between lower-frequency vowels and higher-frequency consonants. The results are compared with basic phonemes; phonemes are the smallest building blocks of human speech. Next, the software takes the data and matches it to a larger group of phonemes, and then to actual words, determining the most likely match.
  3. At the same time as step 2, the software is also taking contextual information into account. This helps the program more accurately predict words that are most likely to be used next. This eliminates the embarrassing mistake of inserting the wrong homonym, like putting Mary instead of merry.
  4. Certain words are put into the most likely sentence structure.
  5. The sentence is transferred to a word processor like Microsoft Word or Notepad.

*Modern Uses: Benefits and Disadvantages*

Voice recognition has many projected uses in the future. Right now however, the industry is concentrating on perfecting this new technology by allowing as many people as possible to access it. Many educators are using voice recognition software to aid in the teaching of disabled students. Adults and children alike who do not have full use of their limbs find speech activated programs the key to making every day life a little easier. Many businesses are also taking advantage in the newest developments in the speech recognition field. The software can make automated systems more user friendly and security precautions more precise. As time goes on, voice recognition programs get easier to use and take less training to operate. However, proper instruction is always an important factor in whether or not the program will be successful with its operator. The major disadvantage to the software is the amount of time it takes to learn how it can be utilized. In today’s busy world, many people want to be able to pop a disk into the computer, speak, and have the optimal output right away. Some day we hope this will be the case with a voice recognition program, but for now technology just isn’t advanced enough.


The light blue area is the amount of disorders associated with repeated trauma. 29,937 of the 281,128 are cases of carpal tunnel syndrome
Carpal tunnel syndrome is the pinching of a nerve in the wrist that causes numbness and tingling. It is also the cause of many work-related injuries and is the reason for many missed days. It is primarily caused by repeated motions of the hand, such as typing or using a computer mouse. Voice recognition software can help alleviate this problem, both before it starts and during recovery. Users do not have to use their hands as often, helping to relieve the pressure.

*So Where Is This Trend Going??*

Some critics say voice recognition is just a trend, and its bound to fade out at any time. Many others feel it is the best thing since sliced bread. The industry is already worth billions of dollars, and it hasn't even started to come standard on most PCs. If something so simple can make life so much easier, its bound to be a big hit, especially with Americans!!



Click the picture for more sources...


Authored by Jocelynn Rego and Jackie O'Neill