Speech recognition and editing system

• US 5,970,460 A
• Filed: 12/05/1997
• Issued: 10/19/1999
• Est. Priority Date: 12/05/1997
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: Lernout & Hauspie Speech Products NV (Intel Corporation)
• Inventors: Rosnow, Harley, Bunce, Erik, Gerard, Douglas
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): ABEBE, DANIEL DEMELASH
• Application Number:
  US08/986,006
• Time in Patent Office: 683 Days
• Field of Search: 704/270, 704/275, 704/178
• US Class Current: 704/278
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/187 Phonemic context, e.g. pronunciation rules, phonotactical constraints or phoneme n-grams
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

The system disclosed herein employs a method for facilitating access to supplemental information relating to text items entered into an editing application program by a speech recognition system, e.g. alternate translations of utterances stored in a dictation event database (DEDB). A speech event database (SEDB) associated with the editing program is maintained so as to contain text items entered by said speech recognition system and context information identifying preceding and following text items. In response to a user initiated change in the editing point, a context list is created characterizing a region of text around the new editing point. The speech SEDB event database is searched for text items matching items in said context list thereby to create a mapping of the list to the SEDB. The speech event database is then used as an index into said dictation event database, e.g. for correcting previously misrecognized utterances.

Claims

1. A method for facilitating access to supplemental information relating to text items entered into an editing application program by a speech recognition system, said supplemental information being stored in a dictation event database (DEDB) maintained by the speech recognition system, said method comprising:

maintaining a speech event database (SEDB) containing text items entered by said speech recognition system;

in response to a user initiated change in the editing point in the editing program, extracting and caching a preselected region of text around the new editing point and creating as a context list characterizing said region;

searching the speech SEDB event database for text items matching items in said context list thereby to create a mapping of the list to the SEDB;

using said speech event database as an index into said dictation event database for accessing said supplemental information.

2. The method of claim 1 wherein said supplemental information includes alternate translations of utterances analyzed by said speech recognition system.

3. The method of claim 1 wherein said speech event database includes context information identifying preceding and following text items.

4. The method of claim 1 further comprising the synthesizing of new items in the speech event database corresponding to context list items for which no match was found.

5. The method of claim 1 wherein said preselected region of text is a text paragraph which includes the new editing point.

6. A method for facilitating access to supplemental information relating to text items entered into an editing application program by a speech recognition system, said supplemental information defining alternate possible translations of utterances stored in a dictation event database (DEDB) maintained by the speech recognition system;

maintaining a speech event database (SEDB) associated with the editing program containing text items entered by said speech recognition system and context information identifying preceding and following text items;

in response to a user initiated change in the editing point in the editing program, extracting and caching a preselected region of that around the new editing point and creating a context list characterizing said region;

searching the speech SEDB event database for text items matching items in said context list thereby to create a mapping of the list to the SEDB;

using said speech event database as an index into said dictation event database for correcting previously misrecognized utterances.

7. The method of claim 6 further comprising the synthesizing of new items in the speech event database corresponding to context list items for which no match was found.

8. The method of claim 6 wherein said preselected region of text is a text paragraph which includes the new editing point.

9. A method for facilitating access to supplemental information relating to text items entered into a word processing application program by a speech recognition system, said supplemental information including alternate translations of dictation events stored in a dictation event database (DEDB)

maintained by the speech recognition system;

maintaining a speech event database (SEDB) associated with the editing program containing text items entered by said speech recognition system;

in response to a user initiated change in the editing point in the editing program, extracting and caching a paragraph and creating as a context list characterizing a region of text around the new editing point;

searching the speech SEDB event database for text items matching items in said context list thereby to create a mapping of the list to the SEDB and synthesizing new items in the SEDB corresponding to context list items for which no match is found;

using said speech event database as an index into said dictation event database for correcting previously misrecognized dictation events.

10. The method of claim 9 wherein said speech event database includes context information identifying preceding and following text items.