Year: 1984
Type: article
Abstract: For speaker-dependent isolated word recognition, a method of pattern matching using the characteristics over a whole word sound as a template has been usually chosen. The method works best when the wordset is not too large, and consists of words with low acoustic inter-word similarity, ie redundant as a series of sounds.
Authors: Shigeyoshi Kitazawa, Masaaki Ishikawa, Shuji Doshita
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Keyword: pattern matching
Open Access status: green
OpenAlex ID: W32010326
