Speech synthesis method for operator assistance telecommunications calls comprising a plurality of text-to-speech (TTS) devices

• US 5,832,433 A
• Filed: 06/24/1996
• Issued: 11/03/1998
• Est. Priority Date: 06/24/1996
• Status: Expired due to Term

• Current Assignee: Verizon Patent and Licensing Incorporated (Verizon Communications Inc.)
• Original Assignee: NYNEX Science & Technology, Inc. (Verizon Communications Inc.)
• Inventors: Basson, Sara, Yashchin, Dina, Kalyanswamy, Ashok, Silverman, Kim
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): Chawan, Vijay B.
• Application Number:
  US08/669,145
• Time in Patent Office: 862 Days
• Field of Search: 379/88, 379/89, 379/67, 704/270, 704/235, 704/260
• US Class Current: 704/260
• CPC Class Codes: H04M 2201/39 using speech synthesis
• CPC Class Codes: H04M 2201/60 Medium conversion
• CPC Class Codes: H04M 2203/354 Reverse directory service
• CPC Class Codes: H04M 3/4931 Directory assistance systems

Abstract

Methods and apparatus are described for providing automated operator services and in particular, a reverse directory assistance service. A calling customer is connected to an automated system that prompts the caller for a listing identifier which is used by the system to retrieve a textual listing corresponding to the listing identifier from a database of textual listings. The textual listing contains a TTS ID which identifies a particular one TTS device from a plurality of TTS devices and the listing is optionally preprocessed and parsed into a plurality of fields which define the listing. The listing text is then sent to the particular one TTS device for text to speech synthesis of the text contained within the listing. The method further includes teaching the system which one TTS device of the plurality of TTS devices, best synthesizes the text contained within the listing and then identifying that one TTS device within the listing so that subsequent synthesis will utilize that TTS device.

Claims

1. In a telecommunications system comprising a database of text entries and a plurality of TTS devices, a method for synthesizing the text contained within the entries comprising the steps of:

establishing a call from a user to a system telephony interface;

prompting the user for an entry identifier;

retrieving a text entry and a TTS identifier associated with said text entry, said text entry corresponding to said entry identifier, the TTS identifier identifying one of the plurality of TTS devices; and

transmitting text included within the retrieved entry to the one of the plurality of TTS devices identified by the TTS identifier.

2. The method according to claim 1, further comprising the steps of:

parsing the text included in the received text entry into one or more distinct fields.

3. The method according to claim 1, wherein said prompting step further comprises the steps of:

prompting the user for a personal identification number.

4. The method according to claim 3, wherein said parsing step parses the retrieved text entry into a name field, an address field and a locality field.

5. The method according to claim 1, wherein said entry identifier is a telephone number.

6. The method according to claim 1, further comprising the step of:

training the system to select which particular one of the plurality of TTS devices is to be used to synthesize the text contained within a particular text entry and modifying the TTS ID associated with that particular entry to be associated with the particular selected one of the plurality of TTS devices.

7. The method according to claim 6, wherein said training step comprises the step of:

prompting the user, after the completion of said transmitting step, for a user provided qualitative assessment of the quality of the speech synthesized by said one of the plurality of TTS devices from said text contained within the particular entry.

8. The method according to claim 7, further comprising the steps of:

parsing the text into one or more distinct fields.

9. The method according to claim 7, wherein said prompting step further comprises the steps of:

prompting the user for a personal identification number.

10. The method according to claim 7, wherein said entry identifier is a telephone number.

11. The method according to claim 8, wherein said parsing step parses the retrieved text entry into a name field, an address field and a locality field.

12. A system for synthesizing speech comprising:

a plurality of text to speech (TTS) devices;

a physical storage device including a database coupled to the plurality of TTS devices, the data base including a plurality of text entries and a plurality of TTS device identifiers, each TTS device identifier corresponding to one of the plurality of TTS devices, each one of the text entries having one of the TTS identifiers associated therewith.

13. The system of claim 12, further comprising:

means for retrieving one of said text entries and the TTS identifier associated with said text entry from the database; and

means for sending the retrieved text entry to the one of the TTS devices identified by the retrieved TTS identifier associated with said retrieved text entry.

14. The system of claim 13, wherein each one of the entries in the database includes:

a text field; and

a TTS device identifier field.

15. The system of claim 14, wherein each one of the entries in the database further includes:

a name field; and

a telephone number field.

16. The system of claim 13, further comprising:

an input device for receiving feedback from a listener; and

means for modifying the TTS identifier in the database associated with a particular one of the text entries in response to feedback from a listener.

17. The system of claim 16, wherein the input device includes a telephone.

18. The system of claim 16, wherein the plurality of TTS devices includes TTS devices with different speech synthesis characteristics.