Method and apparatus for preparing a document to be read by a text-to-speech reader

• US 7,490,040 B2
• Filed: 06/26/2003
• Issued: 02/10/2009
• Est. Priority Date: 06/28/2002
• Status: Active Grant

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: International Business Machines Corporation
• Inventors: Pickering, John B.
• Primary Examiner(s): Hudspeth; David R
• Assistant Examiner(s): Rider; Justin W
• Application Number:
  US10/606,914
• Publication Number:
  US 20040059577A1
• Time in Patent Office: 2,056 Days
• Field of Search: 704/260, 704/271, 704/270
• US Class Current: 704/260
• 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

Abstract

There is disclosed a method and system for preparing a document to be read by a text-to-speech reader. The method can include identifying two or more voice types available to the text-to-speech reader, identifying the text elements within the document, grouping related text elements together, and classifying the text elements according to voice types available to the text-to-speech reader. The method of grouping the related text elements together can include syntactic and intelligent clustering. The classification of text elements can include performing latent semantic analysis on the text elements and characteristics of the available voice types.

Claims

1. A method for automatically marking a document to be read by a text-to-speech reader with voice type identifiers, said method comprising:

identifying two or more voice types available to the text-to-speech reader, each voice type having a corresponding voice type identifier;

identifying text elements within the document, wherein identifying text elements comprises marking gross structural subdivisions of text with a first set of sequenced tags, marking individual paragraphs of the text with a second set of sequenced tags, and marking text elements with a third set of sequenced tags to generate a hierarchical tree identifying the text elements;

grouping similar text elements together, wherein the step of grouping comprises generating one or more clusters according to each identifiable topic of the document, syntactically parsing the document and subsequently performing text mining to determine which text elements in the document are similar, wherein similarity is based upon lexical affinities among the text elements;

classifying the grouped text elements according to voice types available to the text-to-speech reader; and

marking the classified grouped text elements within the document with corresponding voice type identifiers.

2. The method as claimed in claim 1, wherein the step of identifying text elements comprises breaking down the document into elements and separating out the text elements.

3. The method as claimed in claim 1, wherein the step of grouping similar text elements together comprises parsing for structural features of the text elements.

4. The method as claimed in claim 3, wherein the structural features of the text elements include at least one of the position of the text element in the document, the syntax of the text element, and text features within the text element.

5. The method as claimed in claim 3, wherein the step of grouping similar text elements further comprises parsing for thematic features of the text elements.

6. The method as claimed in claim 1, wherein the step of classifying the text elements according to the available voice types comprises finding the best match between the grouped text elements and the characteristics of the voice types.

7. The method as claimed in claim 6, wherein the step of classifying the text elements according to the characteristics of the available voice types comprises identifying similar themes within the text elements and voice types.

8. The method as claimed in claim 6, wherein the step of classifying the text elements according to the characteristics of the available voice types comprises identifying similar intentions within the text elements and voice types.