Personalized text-to-speech services

• US 9,214,154 B2
• Filed: 12/10/2014
• Issued: 12/15/2015
• Est. Priority Date: 06/30/2000
• Status: Expired due to Fees

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Acker, Edmund Gale, Burg, Frederick Murray
• Primary Examiner(s): Armstrong, Angela A
• Application Number:
  US14/565,505
• Publication Number:
  US 20150095034A1
• Time in Patent Office: 370 Days
• US Class Current: 1/1
• CPC Class Codes: G10L 13/02 Methods for producing synthetic speech; Speech synthesisers
• CPC Class Codes: G10L 13/033 Voice editing, e.g. manipulating the voice of the synthesiser
• CPC Class Codes: G10L 19/00 Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis in musical instruments G10H

Abstract

A personalized text-to-speech (pTTS) system provides a method for converting text data to speech data utilizing a pTTS template representing the voice characteristics of an individual. A memory stores executable program code that converts text data to speech data. Text data represents a textual message directed to a system user and speech data represents a spoken form of text data having the characteristics of an individual'"'"'s voice. A processor executes the program code, and a storage device stores a pTTS template and may store speech data. The pTTS system can be used to provide various services that provide immediate spoken presentation of the speech data converted from text data and/or combine stored speech data with generated speech data for spoken presentation.

Claims

1. A method comprising:

receiving, from a sender, a textual message generated by a spoken dialog system;

selecting, via a processor and based on voice characteristics of the sender and the sender speaking a particular set of lines, a speech template from a plurality of speech templates, the speech template comprising information representing characteristics of an individual'"'"'s voice, wherein each speech template in the plurality of speech templates is personalized to the individual and in a distinct language from other speech templates in the plurality of speech templates;

accessing pre-recorded speech from storage corresponding to a first portion of the textual message;

generating variable speech corresponding to a second portion of the textual message; and

merging the pre-recorded speech and the variable speech in an order defined by the speech template.

2. The method of claim 1, wherein selecting of the speech template is further based on an identifier of the sender.

3. The method of claim 2, wherein the individual'"'"'s voice is associated with an individual who is not the sender.

4. The method of claim 1, wherein:

accessing the pre-recorded speech is based on an attribute of the sender, and wherein each of a plurality of speech segments of the pre-recorded speech has characteristics of a unique individual'"'"'s voice.

5. The method of claim 4, wherein the attribute is one of age and gender.

6. The method of claim 1, wherein the speech template represents the characteristics of the voice of one of a parent, sibling, relative, teacher, and friend of the recipient.

7. The method of claim 6, wherein a user receives the spoken version of the textual message with one of a telephone and telephone application programming interface equipped device coupled across a network to a computer.

8. The method of claim 1, wherein the textual message comprises one of an e-mail message and a manuscript text.

9. The method of claim 1, further comprising:

receiving a voice sample from a user; and

generating a user specific speech template for the user based on the voice sample.

10. A system comprising:

a processor; and

a computer-readable storage medium having instructions stored which, when executed by the processor, result in the processor performing operations comprising;

receiving, from a sender, a textual message generated by a spoken dialog system;

selecting, based on voice characteristics of the sender and the sender speaking a particular set of lines, a speech template from a plurality of speech templates, the speech template comprising information representing characteristics of an individual'"'"'s voice, wherein each speech template in the plurality of speech templates is personalized to the individual and in a distinct language from other speech templates in the plurality of speech templates;

accessing pre-recorded speech from storage corresponding to a first portion of the textual message;

generating variable speech corresponding to a second portion of the textual message; and

merging the pre-recorded speech and the variable speech in an order defined by the speech template.

11. The system of claim 10, wherein selecting of the speech template is further based on an identifier of the sender.

12. The system of claim 11, wherein the individual'"'"'s voice is associated with an individual who is not the sender.

13. The system of claim 10, wherein:

accessing the pre-recorded speech is based on an attribute of the sender, and wherein each of a plurality of speech segments of the pre-recorded speech has characteristics of a unique individual'"'"'s voice.

14. The system of claim 13, wherein the attribute is one of age and gender.

15. The system of claim 10, wherein the speech template represents the characteristics of the voice of one of a parent, sibling, relative, teacher, and friend of the recipient.

16. The system of claim 15, wherein a user receives the spoken version of the textual message with one of a telephone and telephone application programming interface equipped device coupled across a network to a computer.

17. The system of claim 10, wherein the textual message comprises one of an e-mail message and a manuscript text.

18. The system of claim 10, the computer-readable storage medium having additional instructions stored which, when executed by the processor, result in the processor performing operations comprising:

receiving a voice sample from a user; and

generating a user specific speech template for the user based on the voice sample.

19. A computer-readable storage device having instructions stored which, when executed by a computing device, result in the computing device performing operations comprising:

receiving, from a sender, a textual message generated by a spoken dialog system;

selecting, based on voice characteristics of the sender and the sender speaking a particular set of lines, a speech template from a plurality of speech templates, the speech template comprising information representing characteristics of an individual'"'"'s voice, wherein each speech template in the plurality of speech templates is personalized to the individual and in a distinct language from other speech templates in the plurality of speech templates;

accessing pre-recorded speech from storage corresponding to a first portion of the textual message;

generating variable speech corresponding to a second portion of the textual message; and

merging the pre-recorded speech and the variable speech in an order defined by the speech template.

20. The computer-readable storage device of claim 19, wherein selecting of the speech template is further based on an identifier of the sender.