Method for electronically generating a spoken message

• US 5,592,585 A
• Filed: 01/26/1995
• Issued: 01/07/1997
• Est. Priority Date: 01/26/1995
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: Lernout & Hauspie NV Speechproducts
• Inventors: Van Coile, Bert, Leys, Steven, Willems, Stefaan
• Primary Examiner(s): MacDonald, Allen R.
• Assistant Examiner(s): EDOUARD, PATRICK NESTOR
• Application Number:
  US08/379,330
• Time in Patent Office: 712 Days
• Field of Search: 381/51, 395/2.76, 395/2.15
• US Class Current: 704/206
• CPC Class Codes: G10L 13/02 Methods for producing synthetic speech; Speech synthesisers
• CPC Class Codes: G10L 13/04 Details of speech synthesis systems, e.g. synthesiser structure or memory management
• CPC Class Codes: G10L 13/06 Elementary speech units used in speech synthesisers; Concatenation rules
• 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

An improved method for generating a spoken message is of the type formed by (i) first recording speech and (ii) then utilizing the recording so as to obtain at least one carrier, each carrier having at least one fixed part and at least one open slot, and then (iii) inserting an argument into each open slot. The improvement involves applying a prosody transplantation technique to the recording in order to obtain a sequence of phonetico-prosodic parameters for each carrier; identifying in each sequence sections of phonetico-prosodic parameters corresponding to the argument of each open slot; and substituting each of the sections by open slot data comprising at least position information indicating the position of each open slot. In a preferred embodiment, the arguments are entered as orthographic or phonetic text and converted to phonetico-prosodic parameters as well, so that the entire spoken message can be synthesized by a phonetics-to-speech system, resulting in enhanced consistency, even when the carriers are generated from the recording of different human subjects.

Claims

1. An improved method for generating a spoken message of the type formed by (i) first recording speech and (ii) then utilizing the recording so as to obtain at least one carrier, each carrier having at least one fixed part and at least one open slot, and then (iii) inserting an argument into each open slot, wherein the improvement comprises:

a) applying a prosody transplantation technique to the recording in order to obtain a sequence of phonetico-prosodic parameters for each carrier;

b) identifying in each sequence sections of phonetico-prosodic parameters corresponding to the argument of each open slot;

c) substituting each of the sections by open slot data comprising at least position information indicating the position of each open slot;

d) assigning to each thus obtained sequence an identifier; and

e) storing the thus obtained sequences with their identifiers in memory.

2. A method according to claim 1, wherein said predetermined message further comprises at least one phrase, said method further comprising:

a) applying a prosody transplantation technique to each of said phrases in order to obtain a further sequence of phonetico-prosodic parameters for each of said phrases;

b) assigning to each of said further sequences each time a further identifier;

c) storing the thus obtained further sequences with their respective further identifier in said memory.

3. A method for electronically generating a spoken message, starting from phonetico-prosodic parameters generated by application of the method according to claim 2, said method comprising:

a) selecting those carriers and phrases composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers and phrases by means of their assigned identifiers;

c) reading said addressed carriers and phrases from said memory;

d) supplying in orthographic form each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating phonetico-prosodic parameters from said orthographic form;

f) filling in said phonetico-prosodic parameters of said arguments in their assigned open slots;

g) transforming said phonetico-prosodic parameters of said carriers and phrases with their arguments into speech.

4. A method for electronically generating a spoken message according to claim 3, wherein said carriers and phrases are concatenated before being transformed into speech.

5. A method for electronically generating a spoken message, starting from phonetico-prosodic parameters generated by application of the method according to claim 2, said method comprising:

a) selecting those carriers and phrases composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers and phrases by means of their assigned identifiers;

c) reading said addressed carriers and phrases from said memory;

d) supplying in phonetic transcription each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating phonetico-prosodic parameters from said phonetic transcription;

f) filling in said phonetico-prosodic parameters of said arguments in their assigned open slots;

g) transforming said phonetico-prosodic parameters of said carriers and phrases with their arguments into speech.

6. A method for electronically generating a spoken message according to claim 5, wherein said carriers and phrases are concatenated before being transformed into speech.

7. A method for electronically generating a spoken message, starting from phonetico-prosodic parameters generated by application of the method according to claim 2, said method comprising:

a) selecting those carriers and phrases composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers and phrases by means of their assigned identifiers;

c) reading said addressed carriers and phrases from said memory;

d) supplying in orthographic form and/or phonetic transcription each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating phonetico-prosodic parameters from said argument;

f) filling in said phonetico-prosodic parameters of said arguments in their assigned open slots;

g) transforming said phonetico-prosodic parameters of said carriers and phrases with their arguments into speech.

8. A method for electronically generating a spoken message according to claim 7, wherein said carriers and phrases are concatenated before being transformed into speech.

9. A method according to claim 2, wherein upon applying said prosody transplantation enriched phonetic transcription is generated.

10. A method for electronically generating a spoken message, starting from phonetico-prosodic parameters generated by application of the method according to claim 9, said method comprising:

a) selecting those carriers and phrases composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers and phrases by means of their assigned identifiers;

c) reading said addressed carriers and phrases from said memory;

d) supplying in orthographic form each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating enriched phonetic transcription from said orthographic form;

f) filling in said enriched phonetic transcription of said arguments in their assigned open slots;

g) transforming said enriched phonetic transcription of said carriers and phrases with their arguments into speech.

11. A method for electronically generating a spoken message, starting from phonetico-prosodic parameters generated by application of the method according to claim 9, said method comprising:

a) selecting those carriers and phrases composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers and phrases by means of their assigned identifiers;

c) reading said addressed carriers and phrases from said memory;

d) supplying in phonetic transcription each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating enriched phonetic transcription from said phonetic transcription;

f) filling in said enriched phonetic transcription of said arguments in their assigned open slots;

g) transforming said enriched phonetic transcription of said carriers and phrases with their arguments into speech.

12. A method according to claim 1, wherein at least one of the following characteristics,

lexical information of the open slot,

syntactical information of the open slot,

intonation model of the open slot,

is determined for each of said sections and each time added to the open slot data of the sequence.

13. A method for electronically generating a spoken message, starting from phonetico-prosodic parameters generated by application of the method according to claim 12, said method comprising:

a) selecting those carriers composing the message to be generated and generating the identifiers assigned to said selected carrierss;

b) addressing in said memory said selected carriers by means of their assigned identifiers;

c) reading said addressed carriers from said memory;

d) supplying in orthographic form each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating phonetico-prosodic parameters from said orthographic form and according to said characteristics;

f) filling in said phonetico-prosodic parameters of said arguments in their assigned open slots;

g) transforming said phonetico-prosodic parameters of said carriers with their arguments into speech.

14. A method for electronically generating a spoken message according to claim 13, wherein said message is formed by at least two carriers which are concatenated before being transformed into speech.

15. A method for electronically generating a spoken message, starting from phonetico-prosodic parameters generated by application of the method according to claim 12, said method comprising:

a) selecting those carriers composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers by means of their assigned identifiers;

c) reading said addressed carriers from said memory;

d) supplying in phonetic transcription each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating phonetico-prosodic parameters from said phonetic transcription and according to said characteristics;

f) filling in said phonetico-prosodic parameters of said arguments in their assigned open slots;

g) transforming said phonetico-prosodic parameters of said carriers with their arguments into speech.

16. A method for electronically generating a spoken message according to claim 15, wherein said message is formed by at least two carriers which are concatenated before being transformed into speech.

17. A method for electronically generating a spoken message, starting form phonetico-prosodic parameters generated by application of the method according to claim 12, said method comprising:

a) selecting those carriers composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers by means of their assigned identifiers;

c) reading said addressed carriers from said memory;

d) supplying in orthographic form and/or phonetic transcription each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating phonetico-prosodic parameters from said argument and according to said characteristics;

f) filling in said phonetico-prosodic parameters of said arguments in their assigned open slots;

g) transforming said phonetico-prosodic parameters of said carriers with their arguments into speech.

18. A method for electronically generating a spoken message according to claim 17, wherein said message is formed by at least two carriers which are concatenated before being transformed into speech.

19. A method for electronically generating a spoken message, starting from phonetico-prosodic parameters generated by application of the method according to claim 1, said method comprising:

a) selecting those carriers composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers by means of their assigned identifiers;

c) reading said addressed carriers from said memory;

d) supplying in orthographic form each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating phonetico-prosodic parameters from said orthographic form;

f) filling in said phonetico-prosodic parameters of said arguments in their assigned open slots;

g) transforming said phonetico-prosodic parameters of said carriers with their arguments into speech.

20. A method for electronically generating a spoken message according to claim 19, wherein said message is formed by at least two carriers which are concatenated before being transformed into speech.

21. A method for electronically generating a spoken message, starting from phonetico-prosodic parameters generated by application of the method according to claim 1, said method comprising:

a) selecting those carriers composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers by means of their assigned identifiers;

c) reading said addressed carriers from said memory;

d) supplying in phonetic transcription each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating phonetico-prosodic parameters from said phonetic transcription;

f) filling in said phonetico-prosodic parameters of said arguments in their assigned open slots;

g) transforming said phonetico-prosodic parameters of said carriers with their arguments into speech.

22. A method for electronically generating a spoken message according to claim 21, wherein said message is formed by at least two carriers which are concatenated before being transformed into speech.

23. A method for electronically generating a spoken message, starting from phonetico-prosodic parameters generated by application of the method according to claim 1, said method comprising:

a) selecting those carriers composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers by means of their assigned identifiers;

c) reading said addressed carriers from said memory;

d) supplying in orthographic form and/or phonetic transcription each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating phonetico-prosodic parameters from said argument;

f) filling in said phonetico-prosodic parameters of said arguments in their assigned open slots;

g) transforming said phonetico-prosodic parameters of said carriers with their arguments into speech.

24. A method for electronically generating a spoken message according to claim 23, wherein said message is formed by at least two carriers which are concatenated before being transformed into speech.

25. A method for electronically generating a spoken message, starting from phonetico-prosodic parameters generated by application of the method according to claim 24, said method comprising:

a) selecting those carriers and phrases composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers and phrases by means of their assigned identifiers;

c) reading said addressed carriers and phrases from said memory;

d) supplying in orthographic form and/or phonetic transcription each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating enriched phonetic transcription from said arguments;

f) filling in said enriched phonetic transcription of said arguments in their assigned open slots;

g) transforming said enriched phonetic transcription of said carriers and phrases with their arguments into speech.

26. A method according to claim 1, wherein upon applying said prosody transplantation enriched phonetic transcription is generated.

27. A method for electronically generating a spoken message, starting from enriched phonetic transcription generated by application of the method according to claim 26, said method comprising:

a) selecting those carriers composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers by means of their assigned identifiers;

c) reading said addressed carriers from said memory;

d) supplying in orthographic form each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating enriched phonetic transcription from said orthographic form;

f) filling in said enriched phonetic transcription of said arguments in their assigned open slots;

g) transforming said enriched phonetic transcription of said carriers with their arguments into speech.

28. A method for electronically generating a spoken message, starting from enriched phonetic transcription generated by application of the method according to claim 26, said method comprising:

a) selecting those carriers composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers by means of their assigned identifiers;

c) reading said addressed carriers from said memory;

d) supplying in phonetic transcription each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating enriched phonetic transcription from said phonetic transcription;

f) filling in said enriched phonetic transcription of said arguments in their assigned open slots;

g) transforming said enriched phonetic transcription of said carriers with their arguments into speech.

29. A method for electronically generating a spoken message, starting from enriched phonetic transcription generated by application of the method according to claim 26, said method comprising:

a) selecting those carriers composing the message to be generated and generating the identifiers assigned to said selected carriers;

b) addressing in said memory said selected carriers by means of their assigned identifiers;

c) reading said addressed carriers from said memory;

d) supplying in orthographic form and/or phonetic transcription each argument to be filled in in said open slots of said selected carriers and assigning each argument to a respective open slot within said selected carriers;

e) generating enriched phonetic transcription from said arguments;

f) filling in said enriched phonetic transcription of said arguments in their assigned open slots;

g) transforming said enriched phonetic transcription of said carriers with their arguments into speech.