But mfcc data I get from Meyda.js is quite different from the mfcc data I get from python_speech_features maybe because I use default setting for Meyda.js and I don't know much about audio signal processing.
I hope some give me hint to get mfcc similar enough to python_speech_features to do deep learning inference which the model we train with Python.
cd web_app
npm install
npm start
cd python_server
python run_keras_server.py
You need to install some libraries to run python server.
npm install --global meyda
pip install python_speech_features
python run_keras_server.py
pip install flask_cors