Voice Recognition and Speech Recognition Software and Vendors Guide

Advantages of Interacting With Computers Through Speech

1. Hands Speech does not require use of physical devices such as keyboards or pointing devices. Computing devices can become more compact as keyboard and mouse pointing devices take a less prominent role. Individuals with physical challenges may also benefit from the use of speech based applications.

2. Eyes Does not require the use of vision. Speech can be performed effectively in low light environments or by persons with low or no vision. Applications for this technology may include uses where the individual needs to keep their eyes on equipment or the environment if navigating a vehicle.

3. History People have been speaking for thousands of years and individuals start speaking at a young age.

4. Distance Speech systems can process speech from individuals that may be at some distance away from the computer.

5. Smaller Wireless and handheld devices are becoming increasingly smaller with corresponding smaller displays and input keys. A voice interface is not constrained by the physical size of the device.

   
   

   What is Speech Recognition?

   
   Speech Recognition is a technology that allows the computer to identify and understand words spoken by a person using a microphone or telephone. The ultimate goal of the technology is to be able to produce a system that can recognize with 100% accuracy all words that are spoken by any person.

   Even after years of research in this area, the best speech recognition software applications still cannot recognize speech with 100% accuracy. Some applications are able to recognize over 90% of words when spoken under specific constraints regarding content and previous training to recognize the speaker's speech characteristics.

   Computer software that understands your speech enables you to have conversations with the computer. These conversations would include you and the computer speaking as commands or in response to events, input, or other feedback.

   Speaking is easier and more intuitive than selecting buttons and menu items. Human speech has evolved over many thousands of years to become an efficient method of sharing information and giving instructions.

   Useful when accessing systems through pay telephones that do not have attached keyboards.

   Description of Speech Recognition (http://www.abilityhub.com/speech/speech-description.htm)
   From AbilityHub Assistive Technology Solutions
   
   
   

   What is continuous speech recognition?

   
   Continuous speech recognition is software that allows the user to speak at a conversational speed. Pauses between words are not needed. Using this software allows you to speak to your computer at a conversational page. The software then transcribes your speech into text as a form of dictation. Continuous speech is not used for command and control of your computer.

   It is software that understands the rules regarding grammar and applies those rules to the speech heard in order to develop a better transcript of the spoken text.

   
   

   What is Discrete voice recognition software?

   
   Discrete speech is spoken to the computer using a brief pause between each word. This allows the computer to attempt to understand each word in isolation. There are some weaknesses in this approach. Since each word is recognized in isolation, there is less accuracy than in systems that use rules of grammar to facilitate recognition.

   ---

   Dictation

   Dictation is a mode of speech recognition in which the user enters data by reading directly to the computer. Dictation mode allows users to dictate memos, letters, and e-mail messages, as well as to enter data using a speech recognition dictation engine. Most recognizers are speaker-dependent. To ensure accurate recognition, a "speaker profile" is created to ensure accurate recognition.

   Dictation software allows you to create documents (email, wordprocessing, forms) by speaking instead of typing. Most dictation software allows you to speak in a natural way (conversational). To work effectively, these systems need to utilize databases of words and sounds (a lexicon). An initial training session is used to help the computer learn to interpret your voice and speech characteristics. This training (called "enrollement") will improve the accuracy of the dictation.

   
   

   Important Things to Consider in Dictation

   
   If you are new to dictation, it may help to begin by using notes or handwritten ideas. These aids will help you to gather your thoughts before speaking. With practice, you should get better at dictating to a machine and maybecome very good at dictating without the use of printed materials.

   When dictating, try not to look at the computer screen since this may distract you from your thoughts and cause you to loose concentration. The computer will attempt to transcribe everything you say. So, make sure you think about it first before saying it.

   Pronounciation (052.htm)
   Speak in a clear voice and pronounce each word. Speech recognition software is not perfect and may make mistakes in the translation of your voice.
   
   
   

   Achieving Good Recognition Accuracy

   
   
   1. Use a noise cancelling microphone
   2. Use a digital microphone instead of an analog one
   3. Make sure background noise is at a minimum
   4. Perform the voice training / user enrollment
   5. Practice at first by reading from a printed source (e.g., newspaper)
   6. Speak clearly and slowly
   7. Think about what you want to say before you say it
   8. Train individual words that are not recognized

   
   

   Commercial Solutions

   
   Philips Digital Dictation System (http://www.dictation.philips.com/index.php?id=51&CC=US)
   Dictation is no longer merely a useful working instrument. It is the commitment to empowering technology for the profitability of your company and for the convenience of the people who work in it.
   
   
   

   Health Care Solutions

   
   EXSpeech (http://www.dictaphone.com/products/exspeech/)
   Enterprise-level speech recognition; input flexibility including phone-based dictation, with transcriptionist editing tools and physician self-completion control.
   
   Enterprise Workstation (http://www.dictaphone.com/products/ews/)
   Web-Based Physician Interface--leverage the power of speech recognition to provide physicians with browser-based once-and-done dictation, report review, and electronic signature capabilities.
   
   PowerScribe for Radiology (http://www.dictaphone.com/products/powerscribe/)
   Departmental speech recognition solution; PowerScribe Workstation is proven, and used by more Radiologists than any other recognition system.
   
   PowerScribe for General Medicine (http://www.dictaphone.com/products/powerscribege/)
   Departmental speech recognition solution, with specialized language models for Emergency Medicine, Pathology, and other medical applications.
   
   ichart Physician Direct (http://www.dictaphone.com/ichart/)
   Lets the smaller practice benefit from the latest digital dictation and speech recognition technology as part of an affordable pay-as-you-go Internet accessible subscription service.
   
   ichart Recognition (http://www.dictaphone.com/products/ichart/)
   Flexible Internet-driven, subscription-based speech recognition built upon the EXSpeech platform.