Need a C++ & Objective C expert to debug Whisper OpenAI Model

Job ID: 35761552

Budget: $50 – $0 CAD

When using the real-time transcription. When the recording time becomes too long and longer, the processing time increases which causes a strain on the phone resources and lag in performance. I want to be able to run live transcript for 3+ hours when I run on a mobile phone without any lags and the application taking up too many resources.

To check what's happening:

Download and run this repo: https://github.com/ggerganov/whisper.cpp

Download Tiny Model using this command:
bash ./models/download-ggml-model.sh tiny.en

I just need for someone to look at the code and advise me on what to do. I can make the changes myself. I need this done quickly tonight.