Here is the synopsis of our sample research paper on Automatic Speech Recognition Systems. Have the paper e-mailed to you 24/7/365.
Essay / Research Paper Abstract
This 14-page paper describes some of the problems inherent in developing automatic speech recognition systems. It discusses the technology (including N-grams, classes and decision trees); barriers; and the problem of accents. Bibliography lists 15 sources.
Page Count:
14 pages (~225 words per page)
File: D0_HVASR.rtf
Buy This Term Paper »
Unformatted sample text from the term paper:
phone calls that have limited scope, such as people calling to make trip reservations. The subject matter is limited (they need to reserve a seat) and the parameters are
few (time and destination). The system can handle these routine calls easily, freeing up human workers for other more complex tasks. Although the technology of
speech recognition systems has been developing for nearly 20 years, the problems associated with it are formidable, and the industry in general has yet to fulfill its promise. This
paper discusses some of the problems associated with the development of the technology. The Technology and the Problems Much of the work currently being done in the speech recognition
field is at the university level, though major companies such as AT&T and IBM are also developing systems (Gallant, 1989). Even though there is a great deal of talk
about the speech recognition system and how it will revolutionize the way we use computers, commercial products are still not widely available; or at least, it appears that most of
the development at the present time is targeted at large installations, such as the reservations systems mentioned above, as opposed to the individual end user, such as a blind person
who needs to be able to "talk" to his computer (Gallant, 1989). Gallant was writing in 1989, and there are more systems available now than there were then; the
point is that they still seem to be targeted more at industry than individuals; that is, weve come to expect to hear "Julie" when we call Amtrak, but its still
unusual to find someone sitting at home using Dragon to dictate to their computer. The systems that are available fall generally into two categories: speaker independent and
...