VocalFlow Clone for Windows

Windows-compatible clone of VocalFlow created for the Hiringg assignment.

This project provides:

• Voice recording from microphone
• Transcription using Deepgram
• Optional text improvement using Grok (xAI-compatible chat API)
• Balance section that shows:
  • Deepgram balance
  • Grok balance / availability

Extra feature requirement covered

1. Show Deepgram balance and Grok balance in the UI.
2. Deepgram key is hardcoded through a config file (config.py) as requested.

Tech stack

• Python 3.10+
• Tkinter (desktop UI)
• Deepgram REST API
• Grok (OpenAI-compatible API style)

Files

• main.py - Main desktop app
• config.py - API keys and constants (contains hardcoded Deepgram key field)
• audio_recorder.py - WAV recording helper
• deepgram_service.py - Deepgram transcription + balance logic
• grok_service.py - Grok text improvement + balance probing
• requirements.txt - Dependencies

Setup (Windows)

1. Open this folder in terminal.
2. Create and activate venv:

python -m venv .venv
.venv\Scripts\activate

3. Install packages:

pip install -r requirements.txt

4. Open config.py and set:

   • DEEPGRAM_API_KEY (required)
   • GROK_API_KEY (optional)

5. Run app:

python main.py

How to use

1. Launch app.
2. Click Refresh Balances.
3. Adjust recording duration.
4. Click Start Recording.
5. Wait for transcript output.
6. Optionally enable Post-process using Grok.

Submission notes

• Do not upload node_modules (not used in this project).
• If zipping this project, exclude:
  • .venv/
  • __pycache__/
  • any local cache files

License

MIT (same as upstream VocalFlow repository)