Method, apparatus, and program for transmitting text messages for synthesized speech

• US 20040143438A1
• Filed: 01/17/2003
• Published: 07/22/2004
• Est. Priority Date: 01/17/2003
• Status: Active Grant

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: International Business Machines Corporation
• Inventors: Moore, Jason Eric, Cabezas, Rafael Graniello, Silvia, Elizabeth
• Granted Patent: US 7,379,872 B2
• Time in Patent Office: Days
• US Class Current: 704/260
• CPC Class Codes: G06F 21/64 Protecting data integrity, e.g. using checksums, certificates or signatures
• CPC Class Codes: G10L 13/033 Voice editing, e.g. manipulating the voice of the synthesiser
• CPC Class Codes: G10L 21/00 Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility G10L19/00 takes precedence

Abstract

A mechanism is provided for authenticating and using a personal voice profile. The voice profile may be issued by a trusted third party, such as a certification authority. The personal voice profile may include information for generating a digest or digital signature for text messages. A speech synthesis system may speak the text message using the voice characteristics, such as prosodic characteristics, only if the voice profile is authenticated and the text message is valid and free of tampering.

Claims

1. A method for generating synthesized speech for a text, the method comprising:

receiving a voice profile including voice characteristic information for an individual from a remote source;

receiving a text message from a remote source;

determining whether the voice profile is signed by a trusted third party;

determining whether the text message is signed using information in the voice profile; and

responsive to the voice profile being signed by a trusted third party and the text message being signed using information in the voice profile, generating synthesized speech for the text message using the voice characteristic information.

2. A method for creating a voice profile by a trusted third party, the method comprising:

verifying the identity of an individual;

obtaining voice characteristics of the individual;

generating a personal voice profile for the individual, wherein the personal voice profile has stored therein profile information including a public key of a private key/public key pair for the individual, an algorithm for signing documents, and voice characteristic information;

generating a digest of the profile information;

encrypting the digest using a private key of the trusted third party to form a digital signature; and

inserting the digital signature into the personal voice profile.

3. The method of claim 2, wherein the trusted third party is a certification authority.

4. A method for transmitting text for synthesized speech, the method comprising:

providing a voice profile including voice characteristic information for an individual, a public key, and an algorithm;

providing a text message to he transmitted;

generating a message digest of the text message using the algorithm;

encrypting the message digest using a private key associated with the public key to form an encrypted digest; and

outputting the text message and the encrypted digest.

5. The method of claim 4, wherein the voice profile further includes at least one of a unique identifier, an expiration, and personal information.

6. The method of claim 4, wherein the voice profile is digitally signed by a trusted third party.

7. The method of claim 4, wherein the step of outputting the text message and the encrypted digest includes one of storing the text message and the encrypted digest on a computer readable medium and sending the text message and the encrypted digest to a recipient over a network.

8. A method for synthesizing speech from a text message, the method comprising:

receiving a voice profile including voice characteristic information for an individual, a public key, and an algorithm;

receiving a text message and an encrypted digest;

decrypting the encrypted digest using the public key to form a decrypted digest;

generating a message digest of the text message using the algorithm; and

responsive to a determination that the decrypted digest and the message digest match, generating synthesized speech for the text message using the voice characteristic information.

9. The method of claim 8, wherein the step of receiving a voice profile includes one of retrieving the voice profile from a computer readable medium and receiving the voice profile from a source over a network.

10. The method of claim 8, wherein the step of receiving a text message and an encrypted digest includes one of retrieving the text message and the encrypted digest from a computer readable medium and receiving the text message and the encrypted digest from a source over a network.

11. The method of claim 8, wherein the voice profile is signed by a trusted third party, the method further comprising:

authenticating the voice profile.

12. The method of claim 8, wherein the voice profile further includes an expiration, the method further comprising:

determining whether the voice profile is expired.

13. The method of claim 12, wherein synthesized speech is not generated for the text message using the voice characteristic information if the voice profile is expired.

14. An apparatus for transmitting text for synthesized speech, the apparatus comprising:

first providing means for providing a voice profile including voice characteristic information for an individual, a public key, and an algorithm;

second providing means for providing a text message to be transmitted;

generation means for generating a message digest of the text message using the algorithm;

encryption means for encrypting the message digest using a private key associated with the public key to form an encrypted digest; and

output means for outputting the text message and the encrypted digest.

15. The apparatus of claim 14, wherein the voice profile further includes at least one of a unique identifier, an expiration, and personal information.

16. The apparatus of claim 14, wherein the voice profile is digitally signed by a trusted third party.

17. The apparatus of claim 14, wherein the output means includes one of means for storing the text message and the encrypted digest on a computer readable medium and means for sending the text message and the encrypted digest to a recipient over a network.

18. An apparatus for synthesizing speech from a text message, the apparatus comprising:

first receipt means for receiving a voice profile including voice characteristic information for an individual, a public key, and an algorithm;

second receipt means for receiving a text message and an encrypted digest;

decryption means for decrypting the encrypted digest using the public key to form a decrypted digest;

digest generation means for generating a message digest of the text message using the algorithm; and

speech generation means, responsive to a determination that the decrypted digest and the message digest match, for generating synthesized speech for the text message using the voice characteristic information.

19. The apparatus of claim 18, wherein the first receipt means includes one of means for retrieving the voice profile from a computer readable medium and means for receiving the voice profile from a source over a network.

20. The apparatus of claim 18, wherein the second receipt means includes one of means for retrieving the text message and the encrypted digest from a computer readable medium and means for receiving the text message and the encrypted digest from a source over a network.

21. The apparatus of claim 18, wherein the voice profile is signed by a trusted third party, the apparatus further comprising:

authentication means for authenticating the voice profile.

22. The apparatus of claim 18, wherein the voice profile further includes an expiration, the apparatus further comprising:

means for determining whether the voice profile, is expired.

23. The apparatus of claim 22, wherein synthesized speech is not generated for the text message using the voice characteristic information if the voice profile is expired.

24. A computer program product, in a computer readable medium, for transmitting text for synthesized speech, the computer program product comprising:

instructions for providing a voice profile including voice characteristic information for an individual, a public key, and an algorithm;

instructions for providing a text message to be transmitted;

instructions for generating a message digest of the text message using the algorithm;

instructions for encrypting the message digest using a private key associated with the public key to form an encrypted digest; and

instructions for outputting the text message and the encrypted digest.

25. A computer program product, in a computer readable medium, for synthesizing speech from a text message, the computer program product comprising:

instructions for receiving a voice profile including voice characteristic information for an individual, a public key, and an algorithm;

instructions for receiving a text message and an encrypted digest;

instructions for decrypting the encrypted digest using the public key to form a decrypted digest;

instructions for generating a message digest of the text message using the algorithm; and

instructions, responsive to a determination that the decrypted digest and the message digest match, for generating synthesized speech for the text message using the voice characteristic information.