Systems and methods for voice synthesis

• US 6,983,249 B2
• Filed: 06/26/2001
• Issued: 01/03/2006
• Est. Priority Date: 06/26/2000
• Status: Expired due to Term

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: International Business Machines Corporation
• Inventors: Sakai, Hideo
• Primary Examiner(s): Young, W. R.
• Assistant Examiner(s): Vo, Huyen X.
• Application Number:
  US09/891,717
• Publication Number:
  US 20020055843A1
• Time in Patent Office: 1,652 Days
• Field of Search: 704/260, 704/270, 704/258, 705/26
• US Class Current: 704/258
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems

Abstract

Systems and methods for voice synthesis are disclosed for providing a synthesized voice message that is consonant with the taste of a customer and a program storage device readable by machine to perform method steps for voice synthesis. In accordance with an order from a customer received via a network, a service provider generates voice synthesis data, based on voice characteristic data for a speaker chosen by the customer, that is produced for a sentence input by the customer, and prepares to deliver the voice synthesis data to the customer. At this time, a transaction number is provided for the order received from the customer, and subsequently, when the transaction number is presented by the customer, the generated voice synthesis data are delivered to the customer. The customer then loads the received voice synthesis data into a device that reproduces the voiced sentence.

Claims

1. A voice synthesis system established between a customer and a service provider who maintains voice characteristic data for multiple speakers, via a network comprising:

a terminal of the customer used by the customer to select a specific speaker from among a list of speakers who are available for the customers selection, wherein the service provider furnishes the list of the speakers via the network, and said terminal used to designate text data for which voice synthesis is to be performed; and

a server of the service provider which employs voice characteristic data for the specific speaker to perform voice synthesis using the text data that is specified by the customer at the terminal to generate voice synthesis data,

whereby the service provider furnishes the customer, together with the list of the speakers, a list of devices into which voice synthesis data can be loaded;

whereby the customer notifies the service provider, via the network, which device was selected from the list; and

whereby the service provider generates voice synthesis data based on the voice characteristic data of the sneaker selected by the customer and loads the obtained voice synthesis data into the device selected by the customer.

2. The voice synthesis system according to claim 1, wherein the server of the service provider assigns a transaction number to the customer;

and wherein, when the transaction number is presented by the terminal of the customer, the server transmits the voice synthesis data to the terminal of the customer.

3. A voice synthesis method employed via a network between a service provider, who maintains voice characteristic data for multiple speakers, and a customer, said method comprising the steps of:

the service provider furnishing a list of the multiple speakers via the network to a remote user;

the customer transmitting to the service provider, via the network, an identity of a speaker that has been selected from the list, and text data for which voice synthesis is to be performed; and

the service provider employing the voice characteristic data for the speaker selected by the customer to perform the voice synthesis using the text data,

whereby the service provider furnishes the customer, together with the list of the speakers, a list of devices into which the voice synthesis data can be loaded;

whereby the customer notifies the service provider, via the network, which device was selected from the list; and

whereby the service provider generates voice synthesis data based on the voice characteristic data of the speaker selected by the customer and loads the obtained voice synthesis data into the device selected by the customer.

4. The voice synthesis method according to claim 3, whereby the service provider assesses a charge for voice synthesis data produced using the voice synthesis, and transmits the voice synthesis data to the customer upon receipt from the customer of payment for the charge.

5. The voice synthesis method according to claim 3, whereby the service provider pays a fee that is consonant with the generation of the voice synthesis data to a person who owns all rights to the voice characteristic data that the service provider holds.

6. A server, which performs voice synthesis in accordance with a request received from a customer connected across a network, comprising:

a voice characteristic data storage unit which stores voice characteristic data obtained by analyzing voices of speakers;

a request acceptance unit which accepts, via the network, a request from the customer that includes text data input by the customer and a speaker selected by the customer from a list of multiple speakers provided by a service provider via a network; and

a voice synthesis data generator which, in accordance with the request received from the customer by the request acceptance unit, performs voice synthesis of the text data based on the voice characteristic data of the selected speaker that are stored in the voice characteristic data storage unit,

whereby the service provider furnishes the customer, together with the list of the speakers, a list of devices into which voice synthesis data can be loaded;

whereby the customer notifies the service provider, via the network, which device was selected from the list; and

whereby the service provider generates voice synthesis data based on the voice characteristic data of the sneaker selected by the customer and loads the obtained voice synthesis data into the device selected by the customer.

7. The server according to claim 6, wherein the voice characteristic data storage unit stores for each speaker, as the voice characteristic data, voice quality data and prosody data.

8. The server according to claim 6, further comprising a price setting unit which sets a price for the voice synthesis data based on the request issued by the customer.

9. A storage device, on which a computer readable program is stored, that permits the computer to perform:

a process for accepting a request from a remote user to generate voice synthesis data for a speaker selected by the remote user from a list of multiple speakers provided by a service provider via a network, wherein the remote user transmitting to the service provider, via the network, an identity of a speaker that has been selected from the list, and text data for which voice synthesis is to be performed, and wherein the service provider employing the voice characteristic data for the speaker selected by the remote user to nerform the voice synthesis using the text data;

a process for, in accordance with the request, generating and outputting a transaction number; and

a process for, upon the receipt of the transaction number, outputting voice synthesis data that are consonant with the request, whereby the service provider furnishes the remote user, together with the list of the speakers, a list of devices into which the voice synthesis data can be loaded;

whereby the remote user notifies the service provider, via the network, which device was selected from the list; and

whereby the service provider generates voice synthesis data based on the voice characteristic data of the speaker selected by the remote user and loads the obtained voice synthesis data into the device selected by the remote user.

10. The program storage device according to claim 9, wherein the program permits the computer to further perform a process which attaches, to the voice synthesis data, verification data for verifying the contents of the voice synthesis data.

11. A storage medium, on which a computer readable program is stored, that permits the computer to perform:

a process, for accepting, for voice synthesis, a request from a remote user that includes text data and a speaker selected by the remote user, from a list of multiple speakers provided by service provider via a network, wherein the remote user transmitting to the service provider, via the network, an identity of a speaker that has been selected from the list, and text data for which voice synthesis is to be performed, and wherein the service provider employing the voice characteristic data for the speaker selected by the remote user to perform the voice synthesis using the text data; and

a process for, in accordance with the request, employing voice characteristic data corresponding to the designated speaker to perform the voice synthesis for the text data; and

whereby the service provider furnishes the remote user, together with the list of the speakers, a list of devices into which voice synthesis data can be loaded;

whereby the remote user notifies the service provider, via the network, which device was selected from the list; and

whereby the service provider generates voice synthesis data based on the voice characteristic data of the speaker selected by the remote user and loads the obtained voice synthesis data into the device selected by the remote user.

12. A program transmission apparatus comprising:

a storage device which stores a program permitting a computer to perform;

a first processor which outputs, to a customer, a list of multiple sets of voice characteristic data stored in the computer;

a second processor which outputs, to the customer, voice synthesis data that are obtained by employing voice characteristic data selected from the list by the customer to perform voice synthesis using text data entered by the customer; and

a transmitter which reads the program from the storage device and transmits the program,

whereby a service provider furnishes the customer, together with a list of multiple speakers from which one speaker can be selected by the customer, a list of devices into which the voice synthesis data can be loaded;

whereby the customer notifies the service provider, via a network, which device was selected from the list; and

whereby the service provider generates voice synthesis data based on the voice characteristic data of the speaker selected by the customer and loads the obtained voice synthesis data into the device selected by the customer.

13. A voice synthesis data storage medium, on which, when a customer connected via a network to a service provider submits a selected speaker chosen from a list of multiple speakers provided to the customer by the service provider via the network, and text data to the service provider, and when the service provider generates voice synthesis data in accordance with the selected speaker and the text data submitted by the customer, the voice synthesis data are stored,

whereby the service provider furnishes the customer, together with the list of the speakers, a list of devices into which the voice synthesis data can be loaded;

whereby the customer notifies the service provider, via the network, which device was selected from the list; and

whereby the service provider generates voice synthesis data based on the voice characteristic data of the speaker selected by the customer and loads the obtained voice synthesis data into the device selected by the customer.

14. A voice output device comprising:

a storage unit, which stores voice synthesis data that are generated by a service provider, who retains in storage voice data for multiple speakers, based on a speaker and text data that are submitted via a network to the service provider; and

a voice output unit which outputs a voice based on the voice synthesis data stored in the storage unit,

whereby the service provider furnishes a customer, together with a list of multiple speakers from which one speaker can be selected by the customer, a list of devices into which the voice synthesis data can be loaded;

whereby the customer notifies the service provider, via the network, which device was selected from the list; and

whereby the service provider generates voice synthesis data based on the voice characteristic data of the speaker selected by the customer and loads the obtained voice synthesis data into the device selected by the customer.

15. A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps for voice synthesis, said method comprising the steps of:

the service provider furnishing a list of the multiple speakers via the network to a remote user;

a customer transmitting to the service provider, via the network, an identity of a speaker that has been selected from the list, and text data for which voice synthesis is to be performed; and

the service provider employing the voice characteristic data for the speaker selected by the customer to perform the voice synthesis using the text data, whereby the service provider furnishes the customer, together with the list of the speakers, a list of devices into which voice synthesis data can be loaded;

whereby the customer notifies the service provider, via the network, which device was selected from the list; and

whereby the service provider generates voice synthesis data based on the voice characteristic data of the speaker selected by the customer and loads the obtained voice synthesis data into the device selected by the customer.