Audio-augmented data keying

• US 5,131,045 A
• Filed: 05/10/1990
• Issued: 07/14/1992
• Est. Priority Date: 05/10/1990
• Status: Expired due to Fees

• Current Assignee: Richard G. Roth
• Original Assignee: Richard G. Roth
• Inventors: Roth, Richard G.
• Primary Examiner(s): Shaw, Dale M.
• Assistant Examiner(s): Doerrler, Michelle
• Application Number:
  US07/521,537
• Time in Patent Office: 796 Days
• Field of Search: 381/41-45, 381/48, 381/51-53, 379/88, 379/97, 379/52, 379/354, 379/355, 395/2
• US Class Current: 704/237
• CPC Class Codes: G10L 15/24 Speech recognition using non-acoustical features
• CPC Class Codes: H04M 11/066 Telephone sets adapted for data transmision

Abstract

Present-day limitations of the conventional touch-tone keypad are overcome permitting alphabetic information to be entered into a distant computer. The caller speaks a speech portion into a telephone handset, and then types out the speech portion on the touch-tone keypad. The computer receiving the call converts the spoken voice information into a form suitable for additional digital processing, as by extracting speech-recognition features from the spoken information. The computer processes the typed numeric string into a list of all the possible combinations of characters it could represent. The extent of correlation between the features of the spoken speech portion and each of the combinations is determined, and the combination having the highest correlation is taken to be the speech portion entered by the user.

Claims

1. A data-keying apparatus for keying data comprising groups of alphabetic characters corresponding to spoken speech portions, comprising:

keypad-data receiving means for receiving a group of numeric keying signals, at least some of said numeric keying signals each corresponding to multiple alphabetic characters;

sound receiving means for detecting sounds corresponding to a spoken speech portion and for extracting features from said sounds;

template store means for storing a multiplicity of speech portion templates, each template indicative of features associated with sounds corresponding to a speech portion and indicative of the spelling thereof;

culling means responsive to said group of numeric keying signals for culling from the speech portion templates in said template store means a subset of speech portion templates such that each speech portion template in the subset has a spelling corresponding to said group of numeric keying signals;

and correlating means responsive to the extracted features for evaluating the correlation between the extracted features and the features of each speech portion template in the subset of speech portion templates, and for identifying the speech portion template in the subset of speech portion templates having the highest correlation with the extracted features.

2. The data-keying apparatus of claim 1, further comprising a speech synthesizer responsive to the correlating means for synthesizing the speech portion corresponding to the speech portion template in the subset of speech portion templates having the highest correlation with the extracted features.

3. The data-keying apparatus of claim 1, further comprising a speech synthesizer responsive to the correlating means for synthesizing a spelling of the speech portion corresponding to the speech portion template in the subset of speech portion templates having the highest correlation with the extracted features.

4. The data-keying apparatus of claim 1 wherein the keypad-data receiving means is a dual-tone multifrequency receiver.

5. The data-keying apparatus of claim 4 wherein the correspondence between numeric keying signals and alphabetic characters is that of a touch-tone telephone keypad.

6. For use with a template store means for storing a multiplicity of speech portion templates, each template indicative of features associated with sounds corresponding to a speech portion and indicative of the spelling thereof, a method for keying data comprising groups of alphabetic characters corresponding to spoken speech portions, comprising the steps of:

receiving via a keypad a group of numeric keying signals, at least some of said numeric keying signals each corresponding to multiple alphabetic characters;

detecting sounds corresponding to a spoken speech portion and extracting features from said sounds;

culling from the speech portion templates in said template store means a subset of speech portion templates such that each speech portion template in the subset has a spelling corresponding to said group of numeric keying signals;

and evaluating the correlation between the extracted features and the features of each speech portion template in the subset of speech portion templates, and identifying the speech portion template in the subset of speech portion templates having the highest correlation with the extracted features.

7. The data-keying method of claim 6, further comprising the step of synthesizing the speech portion corresponding to the speech portion template in the subset of speech portion templates having the highest correlation with the extracted features.

8. The data-keying method of claim 6, further comprising the step of synthesizing a spelling of the speech portion corresponding to the speech portion template in the subset of speech portion templates having the highest correlation with the extracted features.

9. The data-keying method of claim 6 wherein the received keypad data is dual-tone multifrequency data.

10. The data-keying method of claim 9 wherein the correspondence between numeric keying signals and alphabetic characters is that of a touch-tone telephone keypad.

11. A data-keying apparatus for keying data comprising groups of alphabetic characters corresponding to spoken speech portions, comprising:

keypad-data receiving means for receiving a group of numeric keying signals, at least some of said numeric keying signals each corresponding to multiple alphabetic characters;

sound receiving means for detecting sounds corresponding to a spoken speech portion and for extracting features from said sounds;

template generation means for generation of a multiplicity of speech portion templates, each template indicative of a spelling corresponding to said group of numeric keying signals;

and correlating means responsive to the extracted features for evaluating the correlation between the extracted features and the features of each generated speech portion template, and for identifying the generated speech portion template in the multiplicity of speech portion templates having the highest correlation with the extracted features.

12. The data-keying apparatus of claim 11, further comprising a speech synthesizer responsive to the correlating means for synthesizing the speech portion corresponding to the speech portion template in the multiplicity of speech portion templates having the highest correlation with the extracted features.

13. The data-keying apparatus of claim 11, further comprising a speech synthesizer responsive to the correlating means for synthesizing a spelling of the speech portion corresponding to the speech portion template in the multiplicity of speech portion templates having the highest correlation with the extracted features.

14. The data-keying apparatus of claim 11 wherein the keypad-data receiving means is a dual-tone multifrequency receiver.

15. The data-keying apparatus of claim 14 wherein the correspondence between numeric keying signals and alphabetic characters is that of a touch-tone telephone keypad.

16. A method for keying data comprising groups of alphabetic characters corresponding to spoken speech portions, comprising the steps of:

receiving via a keypad a group of numeric keying signals, at least some of said numeric keying signals each corresponding to multiple alphabetic characters;

detecting sounds corresponding to a spoken speech portion and extracting features from said sounds;

generating from the multiple alphabetic characters a multiplicity of speech portion templates;

and evaluating the correlation between the extracted features and the features of each speech portion template in the multiplicity of speech portion templates, and identifying the speech portion template in the multiplicity of speech portion templates having the highest correlation with the extracted features.

17. The data-keying method of claim 16, further comprising the step of synthesizing the speech portion corresponding to the speech portion template in the multiplicity of speech portion templates having the highest correlation with the extracted features.

18. The data-keying method of claim 16, further comprising the step of synthesizing a spelling of the speech portion corresponding to the speech portion template in the multiplicity of speech portion templates having the highest correlation with the extracted features.

19. The data-keying method of claim 16 wherein the received keypad data is dual-tone multifrequency data.

20. The data-keying method of claim 19 wherein the correspondence between numeric keying signals and alphabetic characters is that of a touch-tone telephone keypad.