Information processing apparatus, information processing method, and program

• US 10,025,772 B2
• Filed: 04/21/2014
• Issued: 07/17/2018
• Est. Priority Date: 05/08/2013
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Yamano, Ikuo, Kawano, Shinichi, Tsukahara, Tsubasa, Imoto, Maki
• Primary Examiner(s): Sharma, Neeraj
• Application Number:
  US14/787,595
• Publication Number:
  US 20160110339A1
• Time in Patent Office: 1,548 Days
• Field of Search: 704 9, 704235, 704251, 715257, 715261, 715780, 379 8804, 345654, 345168, 345156
• CPC Class Codes: G06F 40/232 Orthographic correction, e.g. spell checking or vowelisation
• CPC Class Codes: G06F 40/274 Converting codes to words; Guess-ahead of partial word inputs
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

An information processing apparatus may include a control device to determine a plurality of candidate texts to correct a target text which is from an input text string, based on preceding text located at a position preceding the target text and succeeding text located at a position succeeding the target text.

Claims

1. An information processing apparatus comprising:

an input device that receives speech and determines an input text string from the speech by a speech recognition function;

an output device that displays the input text string; and

a control device that determines a plurality of candidate texts to correct a target text which is from the input text string, based on a number value stored in an electronic memory and associated with preceding text located at a position preceding the target text, a number value stored in an electronic memory and associated with succeeding text located at a position succeeding the target text, and a weighting of the number value associated with the preceding text and the number value associated with the succeeding text based on a predetermined ratio, the weighting performed by using a processor to arithmetically apply the predetermined ratio to the number value associated with the preceding text and the number value associated with the succeeding text, such that in determining the plurality of candidate texts one of the value associated with preceding text and the value associated with succeeding text is given greater weight than the other one of the value associated with preceding text and the value associated with succeeding text,

wherein the output device displays an image for use in selecting among the candidate texts.

2. The apparatus of claim 1, wherein the candidate texts are determined from second candidate texts specified based on the preceding text.

3. The apparatus of claim 1, wherein the candidate texts are determined by coupling of the succeeding text to an end of each of the second candidate texts.

4. The apparatus of claim 3, wherein the control device is to calculate an assessment value of the coupling of the succeeding text to the end of each of the second candidate texts.

5. The apparatus of claim 4, wherein the candidate texts are determined based on the assessment values.

6. The apparatus of claim 5, wherein the candidate texts are determined based on some of the assessment values.

7. The apparatus of claim 1, wherein the candidate texts are determined based on a product of an assessment value of second candidate texts specified based on the preceding text and a second assessment value of coupling of the succeeding text to an end of each of the second candidate texts.

8. The apparatus of claim 1, wherein the candidate texts are determined based on pronunciation in relation to the target text.

9. The apparatus of claim 1, wherein the apparatus is in an eye-glass device.

10. The apparatus of claim 9, wherein the candidate texts are determined solely by operation of the control device.

11. An information processing method comprising:

receiving speech by an input device and determining an input text string from the speech by a speech recognition function;

displaying the input text string by an output device;

determining, by a control device, a plurality of candidate texts to correct a target text which is from the input text string, based on a number value stored in an electronic memory and associated with preceding text located at a position preceding the target text, a number value stored in an electronic memory and associated with succeeding text located at a position succeeding the target text, and a weighting of the number value associated with the preceding text and the number value associated with the succeeding text based on a predetermined ratio, the weighting performed by using a processor to arithmetically apply the predetermined ratio to the number value associated with the preceding text and the number value associated with the succeeding text, such that in determining the plurality of candidate texts one of the value associated with preceding text and the value associated with succeeding text is given greater weight than the other one of the value associated with preceding text and the value associated with succeeding text; and

displaying an image by the output device for use in selecting among the candidate texts.

12. A non-transitory recording medium recorded with a program executable by a computer, the program comprising:

receiving speech by an input device and determining an input text string from the speech by a speech recognition function;

displaying the input text string by an output device;

determining a plurality of candidate texts to correct a target text which is from the input text string, based on a number value stored in an electronic memory and associated with preceding text located at a position preceding the target text, a number value stored in an electronic memory and associated with succeeding text located at a position succeeding the target text, and a weighting of the number value associated with the preceding text and the number value associated with the succeeding text based on a predetermined ratio, the weighting performed by using a processor to arithmetically apply the predetermined ratio to the number value associated with the preceding text and the number value associated with the succeeding text, such that in determining the plurality of candidate texts one of the value associated with preceding text and the value associated with succeeding text is given greater weight than the other one of the value associated with preceding text and the value associated with succeeding text; and

displaying an image by the output device for use in selecting among the candidate texts.

13. An information processing system comprising:

an input device that receives speech and determines an input text string from the speech by a speech recognition function;

an output device that displays the input text string;

at least two devices having communication capability of;

an eye-glass device,

a mobile phone device, and

a server device; and

a control device including a plurality of components, in which at least one of the components is included in each of the at least two devices, wherein the components in the at least two devices determine a plurality of candidate texts to correct a target text which is from the input text string, based on a number value stored in an electronic memory and associated with preceding text located at a position preceding the target text, a number value stored in an electronic memory and associated with succeeding text located at a position succeeding the target text, and a weighting of the number value associated with the preceding text and the number value associated with the succeeding text based on a predetermined ratio, the weighting performed by using a processor to arithmetically apply the predetermined ratio to the number value associated with the preceding text and the number value associated with the succeeding text, such that in determining the plurality of candidate texts one of the value associated with preceding text and the value associated with succeeding text is given greater weight than the other one of the value associated with preceding text and the value associated with succeeding text,

wherein the output device displays an image for use in selecting among the candidate texts.