Systems and methods for generating speech of multiple styles from text

1. A method for use in a text-to-speech system comprising a computer-implemented linguistic analysis component operative to generate a phonetic transcription based upon input text, a speech base comprising speech unit recordings associated with a plurality of styles of speech, and at least one computer-implemented speech generation component operative to generate output speech from stored speech unit recordings based at least in part on the phonetic transcription, the method comprising acts of:

(A) receiving, by the linguistic analysis component, input text produced by a text-producing application, wherein the text produced by a text-producing application comprises a speech style indication indicating a style of speech to be output by the text-to-speech system for an associated segment of the input text;

(B) generating, by the linguistic analysis component, a phonetic transcription based at least in part on the input text, the phonetic transcription specifying a first style of speech of the plurality of styles of speech to be output by the at least one speech generation component for the segment of the input text; and

(C) generating, by the at least one speech generation component, output speech based at least in part on the phonetic transcription generated in the act (B), wherein the generating comprises the at least one speech generation component selecting, from the speech unit recordings in the speech base, speech unit recordings associated with a second style of speech of the plurality of styles of speech, the second style of speech being different than the first style of speech, and concatenating the selected speech unit recordings to generate output speech in the first style.

2. The method of claim 1 , wherein the first style is a didactic style, and the second style is a neutral style.

3. The method of claim 1 , wherein the act (C) comprises the at least one speech generation component slowing down an output speech rate and/or inserting at least one pause in the output speech.

4. The method of claim 1 , further comprising an act (D) of generating output speech for another segment of the input text by applying, to the other segment, a statistical model associated with a style of speech specified in the phonetic transcription for the other segment.

5. The method of claim 1 , wherein the act (A) comprises receiving input text comprising a plurality of segments each having an associated speech style indication, at least one of the speech style indications being different than at least one other of the speech style indications, the act (B) comprises generating a phonetic transcription specifying a style of speech to be output for each one of the plurality of segments according to the speech style indication associated with the one segment, and the act (C) comprises generating output speech for each one of the plurality of segments according to the speech style indication associated with the one segment.

6. The method of claim 5 , wherein the plurality of segments constitute a single sentence.

7. The method of claim 1 , wherein the act (B) comprises the linguistic analysis component invoking one or more rules and/or components specific to a style of speech indicated by the speech style indication.

8. A text-to-speech system, comprising:

at least one storage facility storing a speech base comprising speech unit recordings associated with a plurality of styles of speech; and

at least one computer processor programmed to;

receive input text produced by a text-producing application, wherein the text produced by a text-producing application comprises a speech style indication indicating a style of speech to be output by the text-to-speech system for an associated segment of the input text;

generate a phonetic transcription based at least in part on the input text, the phonetic transcription specifying a first style of speech of the plurality of styles of speech to be output for the segment of the input text; and

generate output speech based at least in part on the generated phonetic transcription, the generating comprising selecting, from the speech unit recordings in the speech base, speech unit recordings associated with a second style of speech of the plurality of styles of speech, the second style of speech being different than the first style of speech, and concatenating the selected speech unit recordings to generate output speech in the first style.

9. The text-to-speech system of claim 8 , wherein the first style is a didactic style, and the second style is a neutral style.

10. The text-to-speech system of claim 8 , wherein the at least one computer processor is programmed to generate the output speech by slowing down an output speech rate and/or inserting at least one pause in the output speech.

11. The text-to-speech system of claim 8 , wherein the at least one computer processor is programmed to generate output speech for another segment of the input text by applying, to the other segment, a statistical model associated with a style of speech specified in the phonetic transcription for the other segment.

12. The text-to-speech system of claim 8 , wherein the at least one computer processor is programmed to:

receive input text comprising a plurality of segments each having an associated speech style indication, at least one of the speech style indications being different than at least one other of the speech style indications;

generate a phonetic transcription specifying a style of speech to be output for each one of the plurality of segments according to the speech style indication associated with the one segment; and

generate output speech for each one of the plurality of segments according to the speech style indication associated with the one segment.

13. The text-to-speech system of claim 12 , wherein the plurality of segments constitute a single sentence.

14. The text-to-speech system of claim 8 , wherein the at least one computer processor is programmed to generate the phonetic transcription by invoking one or more rules and/or components specific to a style of speech indicated by the speech style indication in the input text.

15. At least one non-transitory computer-readable storage medium having instructions encoded thereon which, when executed in a computer system, cause the computer system to perform a method comprising acts of:

(A) receiving input text produced by a text-producing application, wherein the text produced by a text-producing application comprises a speech style indication indicating a style of speech to be output by the text-to-speech system for an associated segment of the input text;

(B) generating a phonetic transcription based at least in part on the input text, the phonetic transcription specifying a first style of speech to be output for the segment of the input text; and

(C) generating output speech based at least in part on the phonetic transcription generated in the act (B), wherein the generating comprises selecting, from speech unit recordings in a speech base, speech unit recordings associated with a second style of speech that is different than the first style of speech, and concatenating the selected speech unit recordings to generate output speech in the first style.

16. The at least one non-transitory computer-readable storage medium of claim 15 , wherein the act (A) comprises receiving input text comprising a plurality of segments each having an associated speech style indication, at least one of the speech style indications being different than at least one other of the speech style indications, the act (B) comprises generating a phonetic transcription specifying a style of speech to be output for each one of the plurality of segments according to the speech style indication associated with the one segment, and the act (C) comprises generating output speech for each one of the plurality of segments according to the speech style indication associated with the one segment.