Predicting audience response for scripting

1. A method that is executed by a computer processor on a computing device, the method comprising:

receiving a sub-document unit, wherein a script comprises the sub-document unit;

analyzing content of the sub-document unit and a context of the sub-document unit to extract features of the sub-document unit;

employing a predictive model to predict an anticipated audience response to the sub-document unit based upon the features of the sub-document unit, the anticipated audience response to the sub-document unit being a predicted reaction of an audience responsive to the sub-document unit; and

causing the anticipated audience response to the sub-document unit to be at least one of displayed on a display screen or outputted via a speaker.

2. The method of claim 1 , further comprising determining whether the anticipated audience response to the sub-document unit predicted by the predictive model is one of critical or laudatory.

3. The method of claim 1 , wherein the predictive model is trained using training data that comprises training scripts and sampled audience response data of audience members, wherein the sampled audience response data is responsive to the training scripts, and wherein the sampled audience response data is temporally aligned with the training scripts.

4. The method of claim 3 , wherein the sampled audience response data comprises social media response data of the audience members responsive to the training scripts.

5. The method of claim 3 , wherein the sampled audience response data comprises at least one of search query logs or web page interaction logs of the audience members responsive to the training scripts.

6. The method of claim 3 , wherein the sampled audience response data comprises sensor-collected data of the audience members responsive to the training scripts.

7. The method of claim 1 , wherein the script is at least one of a narrative script or a speech.

8. The method of claim 1 , further comprising receiving a portion of the script in a batch, wherein the portion of the script comprises the sub-document unit.

9. The method of claim 1 , further comprising:

receiving the sub-document unit as the sub-document unit is composed; and

responsive to the sub-document unit being composed:

analyzing the content of the sub-document unit and the context of the sub-document unit to extract the features of the sub-document unit;

employing the predictive model to predict the anticipated audience response to the sub-document unit based upon the features of the sub-document unit; and

causing the anticipated audience response to the sub-document unit to be at least one of displayed on the display screen or outputted via the speaker.

10. The method of claim 1 , further comprising:

segmenting the anticipated audience response to the sub-document unit based on differing values of an audience member characteristic; and

causing the anticipated audience response to the sub-document unit as segmented for the differing values of the audience member characteristic to be at least one of displayed on the display screen or outputted via the speaker.

11. The method of claim 1 , further comprising:

receiving input that specifies a sub-document unit type; and

parsing at least a portion of the script to identify the sub-document unit, wherein the sub-document unit has the sub-document unit type as specified.

12. The method of claim 1 , further comprising receiving input that specifies the sub-document unit, wherein the anticipated audience response to the sub-document unit as specified by the input is predicted and outputted.

13. The method of claim 1 , wherein the content of the sub-document unit comprises one or more of text included in the sub-document unit for delivery by at least one speaker, the at least one speaker specified for the text included in the sub-document unit, or a direction specified in the sub-document unit.

14. A system that predicts a reaction of an audience to a script, comprising:

a reception component that receives at least a portion of the script that comprises a sub-document unit, the sub-document unit being rendered as part of a graphical user interface;

a feature extraction component that extracts features of the sub-document unit based upon content of the sub-document unit and a context of the sub-document unit;

a prediction component that utilizes a predictive model to predict an anticipated audience response to the sub-document unit based upon the features of the sub-document unit, the anticipated audience response to the sub-document unit being a predicted reaction of an audience responsive to the sub-document unit; and

a feedback component that outputs the anticipated audience response to the sub-document unit predicted by the predictive model, the feedback component renders the anticipated audience response to the sub-document unit as part of the graphical user interface.

15. The system of claim 14 being embedded in a word processing system.

16. The system of claim 14 , wherein the predictive model is trained using training data that comprises training scripts and sampled audience response data of audience members, wherein the sampled audience response data is responsive to the training scripts, and wherein the sampled audience response data is temporally aligned with the training scripts.

17. The system of claim 16 , wherein the sampled audience response data comprises social media response data of the audience members responsive to the training scripts.

18. The system of claim 14 , further comprising a recognition component that identifies the sub-document unit from the portion of the script that is received.

19. The system of claim 14 , wherein the prediction component further comprises a filter component that conditions prediction by the predictive model based upon a value of an audience member characteristic.

20. A computer-readable storage medium including computer-executable instructions that, when executed by a processor, cause the processor to perform acts including:

receiving at least a portion of a script during composition;

identifying a sub-document unit from the portion of the script responsive to the sub-document unit being composed;

analyzing content of the sub-document unit and a context of the sub-document unit to extract features of the sub-document unit;

employing a predictive model to predict an anticipated audience response to the sub-document unit based upon the features of the sub-document unit, wherein the predictive model is built based upon social media response data of audience members responsive to training scripts; and

outputting the anticipated audience response to the sub-document unit predicted by the predictive model.