CONTEXTUAL CONVERSION PLATFORM

• US 20110295606A1
• Filed: 05/28/2010
• Published: 12/01/2011
• Est. Priority Date: 05/28/2010
• Status: Active Grant

• Current Assignee: Daniel Ben-Ezri
• Original Assignee: Daniel Ben-Ezri
• Inventors: Ben-Ezri, Daniel
• Granted Patent: US 8,423,365 B2
• Time in Patent Office: Days
• US Class Current: 704/260
• CPC Class Codes: G06F 40/274 Converting codes to words; Guess-ahead of partial word inputs
• CPC Class Codes: G10L 13/08 Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

A contextual conversion platform, and method for converting text-to-speech, are described that can convert content of a target to spoken content. Embodiments of the contextual conversion platform can identify certain contextual characteristics of the content, from which can be generated a spoken content input. This spoken content input can include tokens, e.g., words and abbreviations, to be converted to the spoken content, as well as substitution tokens that are selected from contextual repositories based on the context identified by the contextual conversion platform.

Claims

1. A contextual conversion platform for implementation on a computing device configured to process content of a target for use with a text-to-speech engine, said contextual conversion platform comprising:

a processing module for generating a spoken content input in response to the target, the processing module comprising,

a first filter module for identifying a context identifier in the target, and

a second filter module for delineating portions of the target comprising one or more target terms; and

a repository in communication with the processing module, the repository having entries that are organized according to a prioritized scheme, the entries comprising an identified unit respecting the target terms in the delineated portion and a replacement unit corresponding to the identified unit,

wherein the priority scheme assigns a position for the identified unit amongst a plurality of tables in the repository,

wherein the position is associated with a specificity characteristic that defines a level of priority for identified unit,

wherein the level of priority changes as between the tables, and

wherein the processing module replaces the identified unit in the delineated portion with the replacement unit in accordance with the level of priority to form the spoken content input.

2. A contextual conversion platform according to claim 1 wherein the delineated portions correspond to one or more content identifiers associated with the content of the target.

3. A contextual conversion platform according to claim 1 wherein the context identifier is used to select the tables.

4. A contextual conversion platform according to claim 1 wherein the identified unit comprises one or more variants of the target terms.

5. A contextual conversion platform according to claim 4 wherein the variants comprise the target term in combination with one or more neighboring characters located proximate the target term in the content.

6. A contextual conversion platform according to claim 1 wherein the identified units comprise one or more regular expressions respecting the target terms.

7. A contextual conversion platform according to claim 1 wherein the tables comprises a first table and a second table, and wherein the level of priority decreases from the first table to the second table.

8. A contextual conversion platform according to claim 1 further comprising a text-to-speech engine in communication with the processing module, the text-to-speech engine operatively configured to generate a spoken content from the spoken content input.

9. A computer-implemented method for converting content of a target for use with a text-to-speech engine, said method comprising:

identifying a context identifier from among the content of the target;

separating the content into relevant content and miscellaneous content;

locating a target term in the relevant content;

using the context identifier to identify a substitution unit for the target term; and

generating a spoken content input comprising a replacement unit corresponding to the substitution unit,

wherein the substitution unit is selected from a repository with entries that are arranged in tables in accordance with a priority scheme,

wherein the priority scheme defines a position amongst the tables for the substitution unit, and

wherein the position is associated with a specificity characteristic that defines the level of priority for the identified token.

10. A computer-implemented method according to claim 9 further comprising selecting the table corresponding to the context identifier.

11. A computer-implemented method according to claim 9 further comprising:

performing a first substitution pass and a second substitution pass, wherein the level of priority changes from the first substitution pass to the second substitution pass.

12. A computer-implemented method according to claim 11, wherein the specificity characteristic of the identified units in the first substitution pass is less inclusive than the specificity characteristic of the identified units in the second substitution pass.

13. A computer-implemented method according to claim 9 further comprising performing additional formatting of the spoken content input.

14. A computer-implemented method according to claim 9 further comprising generating spoken content from the spoken content input.

15. A computer-implemented method according to claim 9, wherein the table is associated with the context identifier.

16. A computer-implemented method according to claim 9 wherein the entries comprise at least one variant of the target term.

17. A computer-implemented method according to claim 16 wherein the variant comprises the target term in combination with one or more neighboring characters located proximate the target term in the content.

18. A computer-implemented method according to claim 9, wherein the table comprises at least one regular expression associated with at least one of the terms.

19. A computer-implemented method according to claim 9, wherein the tables comprise a first table and a second table, and wherein the level of priority decreases from the first table to the second table.

20. A computer-implemented method according to claim 9 wherein the content of the target is in a text-based form.