https://github.com/ericz99/interview-gpt
Ace your interviews with your very own "Interviewer Companion"
https://github.com/ericz99/interview-gpt
ai cli gpt-4 llm openai python
Last synced: about 1 year ago
JSON representation
Ace your interviews with your very own "Interviewer Companion"
- Host: GitHub
- URL: https://github.com/ericz99/interview-gpt
- Owner: ericz99
- Created: 2023-11-06T03:37:54.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2023-11-27T17:21:03.000Z (over 2 years ago)
- Last Synced: 2025-06-04T07:48:00.071Z (about 1 year ago)
- Topics: ai, cli, gpt-4, llm, openai, python
- Language: Python
- Homepage:
- Size: 1.82 MB
- Stars: 8
- Watchers: 1
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
## InterviewGPT
WIP
> Your personal interview assistant companion
## Setup
Setting up Development:
> Assuming you have Python 3.9+, and have installed poetry
- `git clone https://github.com/ericz99/interview-gpt.git `
- Run `poetry install` to install all dependency
- Initialize environment `poetry shell`
- How to run?
- You can run with web / server
- Navigate both `/backend/interview_gpt/server` and `/web`
- On backend run `uvicorn main:app --reload`
- On web run `pnpm run dev`
- Or just run the core feature
- Navigate both `/backend/interview_gpt/core`
- Run `poetry run python main.py`
**API Key**
- .env file:
- Create a copy of `.env.template` named `.env`
- Add your OPENAI_API_KEY in .env
## Development
- This is currently project not optimized for performance as of now.
- Working on writing a simple api to create new instance of recorders, and new session
- trying to add VAD (Voice activity detection) instead of relying on RMS
- currently this can only spin up one instance of recorder, and it uses socketio to send data from backend to client
- sometime theres like a delay when speaking to the mic, then couple second later it process the audio data
tldr: not slow nor fast, but works good and transcription is on point. Needs optimized using VAD (Voice activity detection)
## Usage
After running the script, the program will start listening to system audio (eg: Speaker), and will record + create .wav file + transcript for you. After,
the AI will read your .wav to create a transcript then will help you answer question or response back about the transcript.
