A Minimum‐Distance Search Technique and its Application to Automatic Directory Assistance

B. Aldefeld, S. E. Levinson, T. G. Szymanski

Research output: Contribution to journalArticle

Abstract

This paper describes a new search procedure and its application to the problem of obtaining telephone directory information from spoken spelled input. The method obtains its speed from using the concept of equivalence classes, with names classified according to their letter‐by‐letter acoustic similarity. It derives its accuracy from the use of a minimum‐distance criterion for selecting answers. The search procedure finds the name with the minimum distance, usually after only a small fraction of the directory file has been examined. Using an acoustic analyzer with an 80 percent correct recognition rate for individual letters, a 98.6 percent correct recognition rate for names was achieved when the method was applied to a directory of 18,000 entries. On the average, only 1.2 percent of the directory had to be examined for each query. With an input recognition rate of 71 percent for letters, the respective figures were 97.2 percent and 2.8 percent.

Original languageEnglish (US)
Pages (from-to)1343-1356
Number of pages14
JournalBell System Technical Journal
Volume59
Issue number8
DOIs
StatePublished - Oct 1980

Fingerprint

Acoustics
Equivalence classes
Telephone

ASJC Scopus subject areas

  • Engineering(all)

Cite this

A Minimum‐Distance Search Technique and its Application to Automatic Directory Assistance. / Aldefeld, B.; Levinson, S. E.; Szymanski, T. G.

In: Bell System Technical Journal, Vol. 59, No. 8, 10.1980, p. 1343-1356.

Research output: Contribution to journalArticle

@article{e6fb29736f5e4b59b2ac9440ef87292b,
title = "A Minimum‐Distance Search Technique and its Application to Automatic Directory Assistance",
abstract = "This paper describes a new search procedure and its application to the problem of obtaining telephone directory information from spoken spelled input. The method obtains its speed from using the concept of equivalence classes, with names classified according to their letter‐by‐letter acoustic similarity. It derives its accuracy from the use of a minimum‐distance criterion for selecting answers. The search procedure finds the name with the minimum distance, usually after only a small fraction of the directory file has been examined. Using an acoustic analyzer with an 80 percent correct recognition rate for individual letters, a 98.6 percent correct recognition rate for names was achieved when the method was applied to a directory of 18,000 entries. On the average, only 1.2 percent of the directory had to be examined for each query. With an input recognition rate of 71 percent for letters, the respective figures were 97.2 percent and 2.8 percent.",
author = "B. Aldefeld and Levinson, {S. E.} and Szymanski, {T. G.}",
year = "1980",
month = "10",
doi = "10.1002/j.1538-7305.1980.tb03367.x",
language = "English (US)",
volume = "59",
pages = "1343--1356",
journal = "Bell Labs Technical Journal",
issn = "1089-7089",
publisher = "John Wiley and Sons Inc.",
number = "8",

}

TY - JOUR

T1 - A Minimum‐Distance Search Technique and its Application to Automatic Directory Assistance

AU - Aldefeld, B.

AU - Levinson, S. E.

AU - Szymanski, T. G.

PY - 1980/10

Y1 - 1980/10

N2 - This paper describes a new search procedure and its application to the problem of obtaining telephone directory information from spoken spelled input. The method obtains its speed from using the concept of equivalence classes, with names classified according to their letter‐by‐letter acoustic similarity. It derives its accuracy from the use of a minimum‐distance criterion for selecting answers. The search procedure finds the name with the minimum distance, usually after only a small fraction of the directory file has been examined. Using an acoustic analyzer with an 80 percent correct recognition rate for individual letters, a 98.6 percent correct recognition rate for names was achieved when the method was applied to a directory of 18,000 entries. On the average, only 1.2 percent of the directory had to be examined for each query. With an input recognition rate of 71 percent for letters, the respective figures were 97.2 percent and 2.8 percent.

AB - This paper describes a new search procedure and its application to the problem of obtaining telephone directory information from spoken spelled input. The method obtains its speed from using the concept of equivalence classes, with names classified according to their letter‐by‐letter acoustic similarity. It derives its accuracy from the use of a minimum‐distance criterion for selecting answers. The search procedure finds the name with the minimum distance, usually after only a small fraction of the directory file has been examined. Using an acoustic analyzer with an 80 percent correct recognition rate for individual letters, a 98.6 percent correct recognition rate for names was achieved when the method was applied to a directory of 18,000 entries. On the average, only 1.2 percent of the directory had to be examined for each query. With an input recognition rate of 71 percent for letters, the respective figures were 97.2 percent and 2.8 percent.

UR - http://www.scopus.com/inward/record.url?scp=0019068162&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0019068162&partnerID=8YFLogxK

U2 - 10.1002/j.1538-7305.1980.tb03367.x

DO - 10.1002/j.1538-7305.1980.tb03367.x

M3 - Article

AN - SCOPUS:0019068162

VL - 59

SP - 1343

EP - 1356

JO - Bell Labs Technical Journal

JF - Bell Labs Technical Journal

SN - 1089-7089

IS - 8

ER -