ForView - An Interviewing Practice Web Application

ForView is an AI-based web app aimed at non-native speakers to help improve their interview-taking skills. The user will be assigned a random question from a database of common interview questions, and then will record themselves making a short response with their device's camera. Once finished, the app runs the program through a grammar checker API (GrammarBot) and gives the user advice on how to improve the wording of their responses.

Submission for Deltahacks 9 (2022).

Created by Aaron Jumarang, Owen Pan, Jessica Idahosa, and Alex Chen

List of Features:

• Randomly reads and loads a question from a database of interview questions
• Records and saves the user's response to the interview question (using javascript)
• Uses speech_recognition Python library to convert speech in recorded .wav file to text

Planned Features:

• Login authentication system
  • Different files tracking the progress of each individual user
  • Videos stored and associated with each user
• Machine learning implemented to detect facial mannerisms during video recordings
• Detection of fluctuations in voice within newly-converted .wav file
• Implementation of more languages to translate text

Required Dependencies/Packages:

For Python:

1. speech_recognition
2. translate
3. flask
4. pydub
5. ffmpeg