Interview Score Prediction

It is a suggestion-based interview analysis tool that can help companies to score candidates based on lexical and prosodic features.

HiRe is an AI-based interview analysis tool specialized for HR interviews that evaluate candidates based on lexical and prosodic characteristics.

Run Locally

Clone the project

  https://github.com/PALASH-BAJPAI/interview-score-prediction.git

Go to the project directory

  cd App

Start the server

  python main.py

🛠 Technologies

• Python
• Keras
• Tensorflow
• Flask
• Html
• CSS
• JavaScript
• D3.js

Dataset

Audio Dataset

For the audio dataset, we have used the Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS).

Text Dataset

To train our lexical model we used a Stream-of-consciousness data set created during the study of language styles by Pennebaker and King. To earn points they used the Big Five innovation techniques to select the best features to score human goals.

Final Score Dataset

We used MIT dataset from the paper 'Automated Analysis and Prediction of Job Interview Performance', We used this dataset for model which is used to generate final score.

Methodology

Input: An audio and text answer to a question
Output: It would be a score given to each candidate out of 100

Demo

HOME PAGE

PLATFORM PAGE

INTERVIEW PAGE

RESULT PAGE

Feedback

If you have any feedback, please reach out to us at [email protected]