Method for recognizing alphanumeric strings spoken over a telephone network

• US 5,125,022 A
• Filed: 08/10/1990
• Issued: 06/23/1992
• Est. Priority Date: 05/15/1990
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: VCS Industries, Inc. (Koninklijke Philips N.V.)
• Inventors: Schalk, Thomas B., Hunt, Alan K.
• Primary Examiner(s): BROWN, THOMAS
• Application Number:
  US07/566,519
• Time in Patent Office: 683 Days
• Field of Search: 381/42, 381/43, 379/88, 379/91, 379/189, 379/199, 379/89
• US Class Current: 379/88.02
• CPC Class Codes: G07C 9/37 using biometric data, e.g. fingerprints, iris scans or voice recognition
• CPC Class Codes: G10L 15/10 using distance or distortion measures between unknown speech and reference templates
• CPC Class Codes: G10L 17/24 the user being prompted to utter a password or a predefined phrase
• CPC Class Codes: G10L 2015/0631 Creating reference templates; Clustering
• CPC Class Codes: H04M 2201/40 using speech recognition
• CPC Class Codes: H04M 3/382 using authorisation codes or passwords

Abstract

The present invention describes a method for recognizing alphanumeric strings spoken over a telephone network wherein individual character recognition need not be uniformly high in order to achieve high string recognition accuracy. Preferably, the method uses a processing system having a digital processor, an interface to the telephone network, and a database for storing a predetermined set of reference alphanumeric strings. In operation, the system prompts the caller to speak each character of a string, beginning with a first character and ending with a last character. Each character is then recognized using a speaker-independent voice recognition algorithm. The method calculates recognition distances between each spoken input character and the corresponding letter or digit in the same position within each reference alphanumeric string. After each character is spoken, captured and analyzed, each reference string distance is incremented and the process is continued, accumulating distances for each reference string, until the last character is spoken. The reference string with the lowest cumulative distance is then declared to be the recognized string.

Claims

1. A method, using a processing system, for recognizing character strings spoken by a caller over a telephone network, the processing system including a digital processor, means for interfacing to the telephone network and storage means for storing a predetermined set of reference character strings each having at least two characters, comprising the steps of:

(a) initializing a cumulative recognition distance for each of the reference character strings to zero;

(b) prompting the caller to speak a character in a character string to be recognized;

(c) capturing and analyzing the spoken character;

(d) calculating a measure of acoustical dissimilarity between the spoken character and a corresponding character of each of the reference character strings to generate a recognition distance for each of the reference character strings;

(e) incrementing the cumulative recognition distance for each of the reference character strings by the recognition distance generated in step (d);

(f) repeating steps (b)-(e) for each successive character in the character string to be recognized and a corresponding character of each of the reference character strings;

(g) determining which of the reference character strings has a lowest cumulative recognition distance; and

(h) declaring the reference character string with the lowest cumulative recognition distance to be the character string spoken by the caller.

2. The method as described in claim 1 wherein the characters of a reference character string are letters.

3. The method as described in claim 1 wherein the characters of a reference character string are digits.

4. The method as described in claim 1 wherein the characters of a reference character string include both letters and digits.

5. The method as described in claim 1 wherein the step of capturing and analyzing the spoken character uses a speaker-independent voice recognition algorithm and voice recognition class reference data for each character of the string.

6. The method as described in claim 5 further including the step of generating the voice recognition class reference data in an off-line process from a training database of a plurality of training speakers derived over a telephone network.

9. The method as described in claim 1 further including the step of determining whether all of the characters of the string to be recognized have been spoken by the caller prior to step (d).

7. A method, using a processing system, for recognizing character strings spoken by a caller over a telephone network, the processing system including a digital processor, means for interfacing to the telephone network and storage means for storing a predetermined set of reference character strings each having at least two characters, comprising the steps of:

(a) initializing a combined recognition value for each of the reference character strings to zero;

(b) prompting the caller to speak a character in a character string to be recognized;

(c) capturing and analyzing the spoken character;

(d) calculating a measure of acoustical similarity between the spoken character and a corresponding character of each of the reference character strings to generate a recognition value for each of the reference character strings;

(e) incrementing the combined recognition value for each of the reference character strings by the recognition value generated in step (d);

(f) repeating steps (b)-(e) for each successive character in the character string to be recognized and a corresponding character of each of the reference character strings;

(g) determining which of the reference character strings has a highest combined recognition value; and

(h) declaring the reference character string with the highest combined recognition value to be the character string spoken by the caller.

10. The method as described in claim 7 further including the step of determining whether all of the characters of the string to be recognized have been spoken by the caller prior to step (d).

8. A method, using a processing system, for recognizing alphanumeric strings spoken by a caller over a telephone network, the processing system including a digital processor, means for interfacing to the telephone network and storage means for storing a predetermined set of reference alphanumeric strings each having at least two characters, comprising the steps of:

(a) initializing a cumulative recognition distance for each of the reference alphanumeric strings to zero;

(b) prompting the caller to speak a first alphanumeric character in an alphanumeric string to be recognized;

(c) capturing and analyzing the spoken first alphanumeric character;

(d) calculating a measure of acoustical dissimilarity between the spoken first alphanumeric character and a first alphanumeric character of each of the reference alphanumeric strings to generate a recognition distance for each of the reference alphanumeric strings;

(e) incrementing the cumulative recognition distance for each of the reference alphanumeric strings by the recognition distance generated in step (d);

(f) prompting the caller to speak a second alphanumeric character in the alphanumeric string to be recognized;

(g) capturing and analyzing the spoken second alphanumeric character;

(h) calculating a measure of acoustical dissimilarity between the spoken second alphanumeric character and a second alphanumeric character of each of the reference alphanumeric strings to generate a recognition distance for each of the reference alphanumeric strings;

(i) incrementing the cumulative recognition distance for each of the reference alphanumeric strings by the recognition distance generated in step (h);

(j) determining which of the reference alphanumeric strings has a lowest cumulative recognition distance; and

(k) declaring the reference alphanumeric string with the lowest cumulative recognition distance to be the alphanumeric string spoken by the caller.