Voice-dialing system using both spoken names and initials in recognition

• US 5,912,949 A
• Filed: 11/05/1996
• Issued: 06/15/1999
• Est. Priority Date: 11/05/1996
• Status: Expired due to Fees

• Current Assignee: Nortel Networks Limited (Nortel Networks Corporation)
• Original Assignee: Northern Telecom Limited (Nortel Networks Corporation)
• Inventors: Will, Craig A., Chan, Conway
• Primary Examiner(s): Hunter, Daniel S.
• Application Number:
  US08/743,933
• Time in Patent Office: 952 Days
• Field of Search: 379/67, 379/88, 379/89, 379/354, 379/355, 379/88.01, 379/88.02, 379/88.03, 704/251, 704/252, 704/254, 704/255, 704/257, 704/275
• US Class Current: 379/88.03
• CPC Class Codes: H04M 1/271 controlled by voice recognition
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 3/42204 Arrangements at the exchange for service or number selection by voice at the terminal H04M1/27
• CPC Class Codes: H04M 3/4931 Directory assistance systems
• CPC Class Codes: H04Q 2213/1307 Call setup
• CPC Class Codes: H04Q 2213/13093 Personal computer, PC
• CPC Class Codes: H04Q 2213/13096 Digital apparatus individually associated with a subscriber line, digital line circuits
• CPC Class Codes: H04Q 2213/13103 Memory
• CPC Class Codes: H04Q 2213/13106 Microprocessor, CPU
• CPC Class Codes: H04Q 2213/1322 PBX
• CPC Class Codes: H04Q 2213/13377 Recorded announcement
• CPC Class Codes: H04Q 2213/13378 Speech recognition, speech analysis
• CPC Class Codes: H04Q 3/625 Arrangements in the private branch exchange

Abstract

A system for dialing a telephone by voice receives from a user a spoken name corresponding to a telephone number that the user desires to call and at least one initial. The systems uses both forms of speech information to retrieve a stored telephone number that corresponds to a stored name that best matches the spoken name.

Claims

1. A method for dialing a telephone by voice, comprising the steps of:

receiving from a user a speech pattern corresponding to a name in a directory the user intends to call and at least one spoken letter associated with the name, wherein the names in the directory are represented by sound patterns;

retrieving a telephone number corresponding to a name associated with the speech pattern;

converting the sound patterns for the names in the directory to orthographic letters; and

comparing the orthographic letters for the names in the directory with an orthographic representation of the spoken letter.

2. The method of claim 1, wherein the names in the directory are represented by a sequence of orthographic letters, and wherein the retrieving step includes the substeps of:

converting sequences of orthographic letters corresponding to the names in the directory into sequences phonemes; and

comparing the sequences phonemes to the speech pattern to identify a sequence of phonemes for a name in the directory that best matches the speech pattern.

3. The method of claim 1, wherein the names in the directory are represented by a sequence of orthographic letters, and wherein the retrieving step includes the substep of:

comparing the names in the directory with the speech pattern using a phoneme-level representation for the names as an intermediary.

4. A method for providing voice-dialing to users, comprising the steps of:

receiving from a user a speech pattern, the speech pattern indicating a name corresponding to a telephone number that the user intends to call, said speech pattern including a spoken name and at least one letter corresponding to the spoken name;

utilizing the speech pattern to identify a portion of a directory containing different names and corresponding telephone numbers, wherein the names in the directory are represented by sound patterns;

converting the sound patterns for the names in the directory to orthographic letters;

comparing the orthographic letters for the names in the directory with an orthographic representation of the spoken letter;

providing to the user a selection of names from the directory determined to best match the speech pattern; and

initiating a telephone call to one of the telephone numbers in accordance with the user'"'"'s selection of a name.

5. A method comprising the steps of:

receiving from a user a speech pattern corresponding to a name in a directory the user intends to call, wherein the names in the directory are represented by sound patterns;

converting the sound patterns for the names in the directory to orthographic letters;

comparing the orthographic letters for the names in the directory with an orthographic representation of the spoken letter;

presenting the user with a name determined to correspond to the speech pattern; and

receiving from the user an indication as to whether the presented name correctly matches the name the user intends to call, said indication including at least one spoken letter associated with the name the user intends to call.

6. The method of claim 5 further comprising the step of:

retrieving a telephone number corresponding to a name associated with the speech pattern and spoken letter.

7. Apparatus for dialing a telephone by voice, comprising:

a receiver configured to receive from a user a speech pattern corresponding to a name in a directory the user intends to call and at least one spoken letter associated with the name, wherein the names in the directory are represented by sound patterns;

a converter configured to convert the sound patterns for the names in the directory to orthographic letters;

a comparator configured to compare the orthographic letters for the names in the directory with an orthographic representation of the spoken letter; and

retrieving mechanism configured to retrieve a telephone number corresponding to a name associated with the speech pattern.

8. The apparatus of claim 7, wherein the names in the directory are represented by a sequence of orthographic letters, and wherein the retrieving mechanism includes:

a converter configured to convert sequences of orthographic letters corresponding to the names in the directory into sequences phonemes; and

a comparator configured to compare the sequences phonemes to the speech pattern to identify a sequence of phonemes for a name in the directory that best matches the speech pattern.

9. The apparatus of claim 8, wherein the names in the directory are represented by a sequence of orthographic letters, and wherein the retrieving mechanism includes:

a comparator configured to compare the names in the directory with the speech pattern using a phoneme-level representation for the names as an intermediary.

10. Apparatus for providing voice-dialing to users, comprising:

a receiver configured to receive from a user a speech pattern, the speech pattern indicating a name corresponding to a telephone number that the user intends to call, said speech pattern including a spoken name and at least one letter corresponding to the spoken name;

identifying mechanism configured to utilize the speech pattern to identify a portion of a directory containing different names and corresponding telephone numbers name, said names in the directory are represented by sound patterns;

a converter configured to convert the sound patterns for the names in the directory to orthographic letters;

a comparator configured to compare the orthographic letters for the names in the directory with an orthographic representation of the spoken letter;

selection mechanism configured to provide to the user a selection of names from the directory determined to best match the speech pattern; and

an initiator configured to initiate a telephone call to one of the telephone numbers in accordance with the user'"'"'s selection of a name.

11. Apparatus comprising:

a receiver configured to receive from a user a speech pattern corresponding to a name in a directory the user intends to call, said names in the directory is represented by sound patterns;

a converter configured to convert the sound patterns for the names in the directory to orthographic letters;

a comparator configured to compare the orthographic letters for the names in the directory with an orthographic representation of the spoken letter;

presenting mechanism configured to present the user with a name determined to correspond to the speech pattern;

said receiver configured to receive from the user an indication as to whether the presented name correctly matches the name the user intends to call, said indication including at least one spoken letter associated with the name the user intends to call; and

retrieving mechanism to retrieve a telephone number corresponding to a name associated with the speech pattern and spoken letter.

12. The apparatus of claim 11 further comprising:

retrieving mechanism configured to retrieve a telephone number corresponding to a name associated with the speech pattern and spoken letter.