Method for recognizing spoken words

• US 5,454,062 A
• Filed: 12/31/1992
• Issued: 09/26/1995
• Est. Priority Date: 03/27/1991
• Status: Expired due to Term

• Current Assignee: Pronounced Technologies LLC
• Original Assignee: Audio Navigation Systems, Inc.
• Inventors: La Rue, Charles
• Primary Examiner(s): MacDonald, Allen R.
• Assistant Examiner(s): Doerrler, Michelle
• Application Number:
  US07/999,062
• Time in Patent Office: 999 Days
• Field of Search: 395/2, 395/2.4, 395/2.48, 395/2.6, 395/2.63, 395/2.84, 381/29-47
• US Class Current: 704/254
• CPC Class Codes: B60R 16/0373 Voice control in general G10L
• CPC Class Codes: G01C 21/36 Input/output arrangements for on-board computers
• CPC Class Codes: G10L 15/187 Phonemic context, e.g. pronunciation rules, phonotactical constraints or phoneme n-grams

Abstract

A method for identifying any one of a plurality of utterances using a programmed digital computing system, each utterance having an audible form representable by a sequence of speech elements each having a respective position in the sequence. In the computing system, a digital representation corresponding to each of the plurality of utterances is stored a designation respective identifying is assigned to each utterance. A table composed of a plurality of entries is created, each entry being associated with a unique combination of a particular speech element and a particular position in the sequence of speech elements of the audible form of an utterance and storing in each entry the identifying designation of each of the plurality of utterances whose audible form is represented by a speech element sequence containing the particular speech element at the particular position with which that entry is associated; converting an utterance to be identified and spoken by a person into a sequence of speech elements each having a respective position in the sequence; reading each table entry associated with a speech element and position combination corresponding to the combination of a respective position in the sequence of the spoken utterance and the particular speech element at the respective position in the sequence of the spoken utterance; and determining which identifying designation appears most frequently in the entries which have been read in the reading step.

Claims

1. A method for identifying any one of a plurality of words using a programmed digital computing system each word having an audible form representable by a sequence of speech elements, with each speech element having a respective position in the sequence, comprising:

storing, in the digital computing system, a digital representation corresponding to each of the plurality of words and assigning a respective identifying designation to each word;

creating a table composed of a plurality of entries, each entry being associated with a unique combination of a particular speech element and a particular position in the sequence of speech elements of the audible form of a word and storing in each entry the identifying designation of each of the plurality of words whose audible form is represented by a speech element sequence containing the particular speech element at the particular position with which that entry is associated;

receiving a sequence of speech elements spoken by a person and representing one of the plurality of words, and storing representations of the received speech elements and their respective positions in the spoken sequence;

reading each table entry associated with a respective position and speech element combination corresponding to the combination of a respective position in the sequence of received speech elements and the particular received speech element at the respective position in the sequence of spoken speech elements; and

determining which identifying designation appears most frequently in the entries which have been read in said reading step.

2. A method as defined in claim 1 further comprising reproducing the word to which the determined identifying designation is assigned.

3. A method as defined in claim 1 further comprising using the digital representation corresponding to the word to which the determined identifying designation is assigned to control an operation of a machine.

4. A method as defined in claim 1 wherein each speech element is a letter and the audible form of the word consists of the spelling of the word.

5. A method as defined in claim 1 wherein each speech element is a phoneme.

6. A method for identifying any one of a plurality of utterances using a programmed digital computing system, each utterance having an audible form representable by a sequence of speech elements each having a respective position in the sequence, comprising:

storing, in the digital computing system, a digital representation corresponding to each of the plurality of utterances and assigning a respective identifying designation to each utterance;

creating a table composed of a plurality of entries, each entry being associated with a unique combination of a particular speech element and a particular position in the sequence of speech elements of the audible form of an utterance and storing in each entry the identifying designation of each of the plurality of utterances whose audible form is represented by a speech element sequence containing the particular speech element at the particular position with which that entry is associated;

receiving a sequence of speech elements spoken by a person and representing one of the plurality of utterances and storing representations of the received speech elements and their respective positions in the spoken sequence;

reading each table entry associated with a respective position and speech element combination corresponding to the combination of a respective position in the sequence of spoken speech elements and the particular spoken speech element at the respective position in the sequence of spoken speech elements;

determining a plurality of identifying designations that appear with the greatest frequencies in the entries which have been read in said reading step;

reproducing the utterance to which is assigned the determined identifying designation that appears with the highest frequency; and

reproducing, in response to an indication by the person that the reproduced utterance is not the utterance represented by the spoken speech elements, the utterance to which is assigned the determined identifying designation that appears with the second highest frequency.

7. A method as defined in claim 6 wherein each speech element is a letter of a word and each utterance consists of spelling of a word.

8. A method as defined in claim 6 wherein each speech element is a phoneme.

9. A method for identifying any one of a plurality of utterances using a programmed digital computing system, each utterance having an audible form representable by a sequence of speech elements each having a respective position in the sequence, wherein each speech element has at least one identifiable acoustic characteristic and a plurality of the speech elements are substantially identical with respect to the at least one identifiable acoustic characteristic, comprising:

a) storing, in the digital computing system, a digital representation corresponding to each of the plurality of utterances and assigning a respective identifying designation to each utterance;

b) creating a table composed of a plurality of entries, each entry being associated with a unique combination of a particular speech element and a particular position in the sequence of speech elements of the audible form of an utterance and storing in each entry the identifying designation of each of the plurality of utterances whose audible form is represented by a speech element sequence containing the particular speech element at the particular position with which that entry is associated;

c) receiving a sequence of speech elements spoken by a person and representing one of the plurality of utterances and storing representations of the received speech elements and their respective positions in the spoken sequence;

d) reading each table entry associated with a respective position and speech element combination corresponding to each combination of

1) a respective position in the sequence of the spoken speech elements and,

2) at the respective position in the sequence, one of

i) the particular speech element spoken by the person and

ii) each speech element which is substantially identical to the particular speech element with respect to the at least one identifiable acoustic characteristic; and

e) determining which identifying designation appears most frequently in the entries which have been read in said reading step.

10. A method as defined in claim 9 wherein each speech element is a letter of a word and each utterance consists of spelling of a word.

11. A method as defined in claim 9 wherein each speech element is a phoneme.

12. A method for identifying any one of a plurality of utterances using a programmed digital computing system, each utterance having an audible form representable by a sequence of speech elements each having a respective position in the sequence, wherein each speech element has at least one identifiable acoustic characteristic and a plurality of the speech elements are substantially identical with respect to the at least one identifiable acoustic characteristic, comprising:

a) storing, in the digital computing system, a digital representation corresponding to each of the plurality of utterances and assigning a respective identifying designation to each utterance;

b) creating a table composed of a plurality of entries, each entry being associated with a unique combination of a particular speech element and a particular position in the sequence of speech elements of the audible form of an utterance and storing in each entry the identifying designation of each of the plurality of utterances whose audible form is represented by a speech element sequence containing the particular speech element at the particular position with which that entry is associated;

c) receiving a sequence of speech elements spoken by a person and representing one of the plurality of utterances and storing representations of the received speech elements and their respective positions in the spoken sequence;

d) reading each table entry associated with a respective position and speech element combination corresponding to each combination of

1) one of a respective position in the sequence of the spoken speech elements and a position in the sequence adjacent the respective position, and

2) the particular spoken speech element at the respective position in the sequence; and

e) determining which identifying designation appears most frequently in the entries which have been read in said reading step.

13. A method as defined in claim 12 wherein each speech element is a letter of a word and each utterance consists of spelling of a word.

14. A method as defined in claim 12 wherein each speech element is a phoneme.