Trusted since 2008
LoginRegister
Study resources on Kenyaplex

Get ready-made curriculum aligned revision materials

Exam papers, notes, holiday assignments and topical questions – all aligned to the Kenyan curriculum.

Browse exam papers
Or go directly to:
Class notes Holiday assignments Topical questions Other resources

List the limitations of Speech Input devices.

List the limitations of Speech Input devices.

Answers

Kavungya
1. Homophones – some words have same sounds.
2. Word separation.
The speech must be “clipped” i.e. each word must be followed by a short period of silence. This enables the device to recognize the end of each word. If this is not done, the device might not be able to tell the difference between ‘Command” and “Come and”, depending on the accent of the speaker.
3. Speaker variability.
The speed, tone (quality of sound), accent, loudness and pronunciation of an individual speaker can vary
Voice input is complex to develop, and it does not take care of speech related problems such as accents and tone. This implies that the device must learn the unique speech of an individual.
4. Limited vocabulary understood & the no. of speakers they can recognize.
Most speech systems can recognize a limited, standard vocabulary of spoken words. The simplest SRD can recognize the voice of only one speaker. .
The device must also be ‘taught’ the voice patterns of the speaker. For example, if a device can store a vocabulary of 25 words, the user will speak the 25 chosen words into the device. The device then stores the patterns for the spoken words. Later, when the user speaks a word, the device compares the pattern of the word spoken with those patterns it had previously stored. If a match is found, the word is recognized. If not, the speaker may be requested to repeat the word or use a synonym (a word that means almost the same thing).
5. The response rates of these devices are still relatively slow.
When you compare the no. of words in English & the total no. of words that can be said at a given point, show that speech recognition is slow.
6. Speech input is complex to develop & is still at the early stages of development.
Kavungya answered the question on April 2, 2019 at 09:16

Answer Attachments

Next: Read the poem below and answer the questions that follow.

Previous: Read the comprehension below and then answer the questions that follow

View more Computer Studies Questions and Answers  |  Return to Questions Index

Exams With Marking Schemes

Related Questions