-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
prompt engineering; output to two text files (#19)
- Input the full transcription of the presentation and apply prompt engineering to generate questions targeted for each section and overall suggestions on how to improve the content of the presentation. - The output contains two styles of questions to reflect avatars with different expertise. - *Note: Since we are still integrating the components this week, I stored the output from people with different expertise in separate text files ‘Response1.txt’ and ‘Response2.txt’.*
- Loading branch information
1 parent
a7e0d1a
commit 545d06a
Showing
14 changed files
with
201 additions
and
22 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file was deleted.
Oops, something went wrong.
Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.
Oops, something went wrong.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,34 @@ | ||
Prompt: | ||
Given the full transcription for a presentation, do the following: | ||
1. Read the full transcript and divide it into separate sections (e.g., motivation, overview, related work). | ||
2. Pretend you are a professional well-versed in the topic. | ||
3. Ask questions targeted for each section. Make sure to include at least one question for each section. Make sure the questions fit your character. Make sure each question is no more than two sentences. Do not contain the word 'Novice' in the question. | ||
4. List at least two areas of improvement in the presentation. Make sure the suggestions fit your character. Make sure each suggestion is no more than two sentences. Do not contain the word 'Novice' in the suggestion. | ||
5. Pretend you are a novice who is not well-versed in the topic. | ||
6. Ask questions targeted for each section. Make sure to include at least one question for each section. Make sure the questions fit your character. Make sure each question is no more than two sentences. Do not contain the word 'Professional' in the question. | ||
7. List at least two areas of improvement in the presentation. Make sure the suggestions fit your character. Make sure each suggestion is no more than two sentences. Do not contain the word 'Professional' in the suggestion. | ||
8. Output your answer in the following format: | ||
''' | ||
Professional: | ||
Questions: | ||
Section <section name> | ||
1. ... | ||
2. ... | ||
Section <section name> | ||
1. ... | ||
2. ... | ||
Suggestions: | ||
1. ... | ||
2. ... | ||
Novice: | ||
Questions: | ||
Section <section name> | ||
1. ... | ||
2. ... | ||
Section <section name> | ||
1. ... | ||
2. ... | ||
Suggestions: | ||
1. ... | ||
2. ... | ||
''' |
Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.
Oops, something went wrong.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,16 @@ | ||
Questions: | ||
Section Motivation | ||
1. Can you elaborate on the specific limitations encountered when practicing presentations in front of a wall or mirror that Deck Talk aims to overcome? | ||
2. How does the inclusion of AI-powered audiences in virtual reality fundamentally enhance the effectiveness of practice sessions compared to traditional methods? | ||
|
||
Section Overview | ||
1. You mentioned three core concepts of Deck Talk; could you discuss how each concept individually contributes to improving presentation skills? | ||
2. Regarding interactivity with the audience, what types of feedback can the presenter expect to receive from the AI? | ||
|
||
Section Related Work | ||
1. How does Deck Talk's feature set compare to Virtual Speech with regard to real-time interactivity and feedback accuracy? | ||
2. Given the competitive landscape, what are the commercial viability and sustainability plans for Deck Talk considering it will be offered free? | ||
|
||
Suggestions: | ||
1. Consider conducting a side-by-side comparative study detailing user improvement using Deck Talk versus traditional practice methods to underline the app's effectiveness. | ||
2. Include testimonials from early users to add authenticity and real-world applicability in your next presentation. |
Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.
Oops, something went wrong.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,17 @@ | ||
Questions: | ||
Section Motivation | ||
1. Why do people practice talks in front of a mirror and what issues do they face? | ||
2. How might practicing in a virtual environment be different from practicing in front of a mirror? | ||
|
||
Section Overview | ||
1. What exactly do you mean by ‘virtual avatars’? | ||
2. Could you explain a little more about how interacting with the AI audience works? | ||
|
||
Section Related Work | ||
1. What is the main difference between Deck Talk and other similar apps like Virtual Speech? | ||
2. Why is it important that Deck Talk is free? | ||
|
||
Suggestions: | ||
1. Could you include more simple examples or comparisons to help understand how the virtual reality environment works? | ||
2. Maybe add some pictures or screenshots of the app in action, so it's easier to visualize what you are talking about. | ||
``` |
Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.
Oops, something went wrong.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,20 @@ | ||
Full transcription: | ||
We present deck talk a virtual reality app that enables users to practice giving talks in virtualized environments to an interactable AI powered audience. | ||
Presentation it's common in both academic and professional settings | ||
To refine the presentation people often practice giving talks in front of a wall or mirror | ||
Where are limited amount of feedback can be obtained | ||
How can we make practice talks more effective and closer to real life? | ||
We designed deck talk based on three core concepts. | ||
First the user should be able to choose from a diverse set of environments to give toxin | ||
Second | ||
We are just taking virtual avatars should be present in the scene. | ||
To mimic real audiences. | ||
Third the user should be able to interact with the audience via question answering and receive feedback about the presentation | ||
The company virtual speech has developed a similar application with virtual reality and AI. | ||
Including varied Workplace settings | ||
Virtual Interactable avatars. And AI generated ratings. | ||
However, virtual speech requires the conversion of presentation sites into PDFs | ||
And it costs $400 a year | ||
In contrast we allow users to bring the presentation into the virtual environment | ||
The web browser and are planning to offer the app for free by open sourcing the code. | ||
In addition our app enables users to improve their general presentation skills by practicing against AI generated presentation sides. |
File renamed without changes.