A Python based Web & REST API built on Flask web framework. 

OBJECTIVES: 
Speaker registration is used for recognizing or identifying from persons individual sound by machine learning. 

Python 3.7+

This application requires an FFMPEG MP3 to WAV file conversion to be installed at system level if you are planning to support MP3 audio files.

Step 2: 16000 kHz 
Step 3: Mono type
To record from your computer use # Audacity tool with above specifications. From mobile and tabs by default MP3 audio clip will be generated.

Step 1: python3 -m venv env
Step 2: source env/bin/activate
Step 3: run pip install -r requirements.txt

Step4: flask run app.py

Navigate to http://host-name:port/upload path to upload wav audio file.

Step 1: Choose .wav file at present minimum 45 seconds of audio clip is required to get good accuracy.
Step 2: Enter any unique in the user name input field
Step 3: Click on Upload

Navigate to http://host-name:port/authentication-upload path to test.
Step 1: Choose .wav file at present minimum 15 seconds of audio clip </br>is required to get good accuracy.
Step 2: Click on Test button