Method for recognizing spoken identifiers having predefined grammars

• US 6,865,532 B2
• Filed: 09/19/2001
• Issued: 03/08/2005
• Est. Priority Date: 09/19/2001
• Status: Expired due to Term

• Current Assignee: CDN Innovations, LLC (IP Investments Group LLC)
• Original Assignee: Mitsubishi Electric Research Laboratories, Inc. (Mitsubishi Electric Corporation)
• Inventors: Anderson, David B.
• Primary Examiner(s): McFadden, Susan
• Assistant Examiner(s): Vo, Huyen X.
• Application Number:
  US09/956,307
• Publication Number:
  US 20030061045A1
• Time in Patent Office: 1,266 Days
• Field of Search: 704/9, 704/252, 704/251, 704/209, 704/270.1, 707/5, 379/88.01
• US Class Current: 704/257
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

A method for selecting and recognizing spoken identifiers first defines a phrase having word slots. The word slots are arranged in the phrase in a predetermined order and according to a predetermined grammatical structure of a target language. A set of unique words selected from the target language is assigned to each word slot in the phrase according to the grammatical structure. Then, a unique identifiers can be generated by selecting one word from each set for each slot for each identifier such that a concatenating of the selected words in the predetermined order form the unique identifier.

Claims

1. A method for selecting and recognizing spoken identifiers, comprising:

defining a phrase having a plurality of word slots, the plurality of word slots arranged in the phrase in a predetermined order and according to a predetermined grammatical structure of a target language;

associating a set of unique words with each word slot, the words in each set selected from the target language according to the grammatical structure;

generating a plurality of unique identifiers by selecting one word from each set for each word slot for each identifier such that a concatenating of the selected words in the predetermined order form the unique identifier.

2. The method of claim 1 wherein the words for each unique identifier are selected by an associated user, and further comprising:

verifying the unique identifiers using a common storage device.

3. The method of claim 1 wherein each unique identifier is a logical identifier, and further comprising:

associating each logical identifier with a physical identifier.

4. The method of claim 1 wherein a plurality of target languages are associated with the phrase, each target language having associated sets of words and grammatical structure different than all other target languages.

5. The method of claim 4 wherein a particular unique identifier is a different concatenation of different words in each target language.

6. The method of claim 5 wherein each different concatenation of different words for the particular unique identifier maps to a single numeric identifier.

7. The method of claim 1 further comprising:

recognizing a particular unique identifier with a speech recognizer.

8. The method of claim 7 further comprising:

applying forward error correction to the recognized unique identifier.