Comparison Chart of Speech-to-Text Systems

	Description	Equipment	Training	Verbosity (pages/hour)
CART	Provider types nearly every spoken word. This steno-based approach is described as verbatim.	Eight-key steno machine is used to record court proceedings and specialized software is used to create a real-time text display.	Requires two to four years. Associate or Bacalaureate degrees offered. List of approved reporting programs at the National Court Reporters Association website. www.ncraonline.org	15 - 20
Note: In Meaning- for- Meaning Systems (e.g., C-Print® and TypeWell®) the service provider uses a laptop computer and proprietary software to create a real-time text display on the same or a second laptop. The laptops are connected via cable or wirelessly. The service provider types only the meaning of the spoken words; repetitions, interjections and other extraneous material is ignored.
C-Print®	Provider interprets (i.e. listens for content) what the speaker says and using a standard keyboard and phonetic-based abbreviation system delivers a meaning-for-meaning text display.	C-Print® software is used to produce text on a laptop computer or other display monitor.	Requires an initial (approximately) 60 hours of training in the abbreviation system and text-condensing strategies. Additional training is recommended for the provider to increase real-time skills before entering the classroom.	6 - 10
TypeWell®	Provider interprets (i.e. listens for content) what the instructor says and using a standard keyboard and spelling-based abbreviation system delivers a meaning-for-meaning text display.	Laptop computer is used with TypeWell® software	Requires an initial (approximately) 60 hours of training in the abbreviation system and text-condensing strategies. Additional training is recommended for the provider to increase real-time skills before entering the classroom.	6 - 10
Speech Recognition	A service option that enables a speech-to-text service provider to create a real-time text display using speech recognition software. The process often is described as voice-to-text.	Provider speaks into a microphone and the software translates the speech into text. A computer keyboard may be utilized during this process to change a word or phrase not recognized by the software or recognized incorrectly.	Requires extended voice training to achieve a high rate of accuracy. While the initial voice training may take anywhere from five to 30 minutes, it takes significantly more time to ensure quality output.	15 - 20

Same Speech, Three Different Transcripts

From dss.jsu.edu/pp/m4m_files/frame.html
(Excerpt taken from PEPNet 2008 Conference presentation by Cindy Camp, Judy Colwell and Pam Francis titled, "Meaning-for-Meaning in Speech-to-Text Services: A Better Understanding."

Verbatim
Here is a cross section of the Great Pyramid. One of the things they did, is you have all of this work going on here. Here is a chamber cut into the bedrock. All the way down here. What people can't figure out is: How did they get the fresh air -- this is like 300 feet or more down this shaft into this big chamber that was cut into the bed-rock. How did they get fresh air down in there? Because you are going to have people down there working, breathing, and exhaling. After a while, they would use up all the oxygen. So, how do they get fresh oxygen down there for those people. That is one problem.

Meaning-for-Meaning
Here is a cross section of the Great Pyramid. One of the things they did, you have all this work going on, a chamber built into the bedrock. How did they get the fresh air 300 feet or more down this shaft into this chamber? Because you will have people down there working, breathing and exhaling, and after a while use up all the oxygen.

Meaning-for-Meaning
Here is a cross section of the Great Pyramid. There is all this work going on. There is a chamber cut into the bedrock. How did they get fresh air in there? People are breathing and exhaling and they would use all the oxygen.

Picture of a man using Speech-to-Text transcription machine

A Student Says:

"Using C-Print® in high school had a positive impact on my performance in the classroom. C-Print® enabled me to follow more closely what was happening during class as well as seeing what comments my fellow classmates were making. This allowed me to be more involved in class discussions and enabled me to relate better to my peers. As it did in high school, C-Print® continues to do so in college, allowing me to tackle my studies with the same gusto as my hearing peers."
Alexandra Johnson, Student, Rochester Institute of Technology

Professionals Say:

"With verbatim, it is English, my native language and it is so fast! I like hearing what I can hear, but following the captions to fill in the gaps." Pietrina Probst, deaf professional

"I have used CART, my preferred communication access mode, from its inception. Because it is verbatim, I know that I won't miss anything." Heidi Adams, PEPNet Outreach Specialist