Medical Transcription Editing Industry

Card Set Information

Medical Transcription Editing Industry
2013-10-23 09:30:38
Speech Recognition – Timeline

Speech Recognition – A Timeline
Show Answers:

  1. in the beggining
    Interest and research in speech recognition has been around for a long time.  In the 1930s there was work being done by AT&T in creating electronic speech using a “synthesized” voice.  For many years, however, the technology did not make any great strides because of the lack of power and speeds of computers and the variable nature of the spoken word.
  2. 1960's
    1960sIBM develops and demonstrates the IBM “Shoebox.” This device responded to 16 spoken words including numeric digits from 0 to 9. Much like today’s SRT devices, the Shoebox was operated by speaking into a microphone and the device would convert voice sounds to electrical impulses.
  3. 1970's
    1970sThe U.S. Defense Advanced Research Projects Agency (DARPA) funds projects for Speech Understanding Research.  Apparently, the United States military saw the potential use for the ability to convert speech to text.  It wasn’t until the late 1970s that the technology became commercially available, with the technology costing up to $100,000.
  4. 1980's
    1980sSpeech recognition technology meets with a fork in the road and splits into two different directions: Call center speech recognition systems and speech-to-text applications.

    It is in the 1980s that SRT begins to gain recognition when Dragon Systems demonstrates a PC-based application that is capable of recognizing 8,000 words.
  5. 1990's
    1990s Continuous speech recognition technology hits the shelves.  This is different from earlier models because it recognizes speech by phrases and takes the context of the speech into account.  This increases accuracy and allows the dictator to speak more naturally. New SRT models now have a vocabulary of up to 23,000 words, and continuous speech systems become available for desktop computers and handheld devices.
  6. 2000 and beyond
    2000 and BeyondSpeech recognition technology is becoming more diverse and companies are successfully tailoring the technology to different applications.  It is available in industry-specific forms for court reporting and medical transcription.  It is increasingly available for entertainment, communication, disabled users, and wireless devices.