This is the first work package part of a project to build a user-friendly and highly interactive online meeting assistant tool. The project will be divided into small work packages that will be built on top of each other. The purpose of work package #1 is to build a web interface that will only have the following functionalities:
- Function to load a locally stored file that contains a text structured in JSON format created by a ASR (Automated Speech Recognition) engine. The engine splits the text in multiple entries. One entry is spoken by one person and contain one or more sentences.
- Once the file is loaded, the content should be visible on a section of the web page (maximum 40% of the web page area to keep room for other functionalities).
- On the same page, the user should be able to interactively activate a "create sections" feature that would allow him/her to select the algorithm which will be used to break the displayed text into different sections, each section containing multiple entries. Only one algorithm will be required for this work package: split the text based on a fixed number of JSON entries. The number of entries must be manually inputted by the user. Once this feature is activated and algorithm selected, the text should be reformatted, in the same section that was used to list the text before, highlighting the different sections. E.g., if a text has 100 entries and the user selects a granularity of 20 entries, we should clearly be able to identify the 5 text sections generated. The way the highlighted sections should be structured should allow for further manual manipulation of each section (drag and drop of text from another section, manual add/remove of text etc.). These features will be implemented in later phases.
- If the "create sections" feature is disabled, the text window should show the initial text, not formatted per sections.
- All the functionalities described above should be part of a single web interface.
- Unless specifically mentioned, the developer is free to decide how the design and functionality should look like.
- Manual manipulations on text sections that are created via the "create sections" functionality.
- Adding more algorithms to create the sections.
- Ability to load/transform the text as unformatted text.
- Automatically highlight key words and sentences based on pre-defined algorithm. Ability to manually override the highlight algorithm in very easy and dynamic way.
- What technologies do you plan to use to build this application? Please explain why, considering the overall context of the project.
- What kind of similar experience do you have?
- How much time it would take you to build this?
- What questions do you have for us, that would help you deliver the best results for this work package?