Polymath

This is a fork of the original Polymath project with several aims.

1. Clean up and organizing the original codebase to make it easier to read

2. Utilize a human readable database

3. Create a more intuitive cli

How it works

This fork utilizies tinydb to create a Json database and fire to create the cli

Folder structure

/polymath
	cli.py 
	/internals
	/content          
		/db
			db.json
			/FrequencyFrames
		/ytdl_content
		/processed
		/stems 

content

• The content dir and subdirectories are autogenerated if they do not already exist

• You should see this message the first time you run any command

Created folder at:
        C:\Users\...\polymath\content

Created folder at:
        C:\Users\...\polymath\content\processed
		
Created folder at:
        C:\Users\...\polymath\content\stems
		
Created folder at:
        C:\Users\...\polymath\content\db
		
Created folder at:
        C:\Users\...\polymath\content\db\frequencyFrames
		
Created folder at:
        C:\Users\...\polymath\content\ytdl_content

db

• The db directory contains a db.json file which is used as the main database for this project.

• The frequencyFrames directory within the db directory contains frequency frames for each audio file added to the database as a text file

ytdl_content

• This directory contains all the downloaded audio files from youtube-dl

processed

• Before a song is added to the database it is cleaned and processed by converting it to a WAV and resampling it a samplerate of 44100 if it is not already

• This directory contains the processed audio files.

stems

• This directory contains the separated stems generated the processed audio filess

cli.py

• This is the main script to run the command line interface (CLI).

The database

• When a song is added to the database it will populate a record that looks like this :

    "1": {
      "id": "song_name_ID",
      "songName": "song_name",
      "pathToFile": "C:\\Users\\...\\polymath\\content\\processed\\song_name_ID.wav",
      "features": {
        "id": "song_name_ID",
        "tempo": "129.19921875",
        "duration": "180.03591836734694",
        "timbre": "-10.74387",
        "timbre_frames": "[[...]]",
        "pitch": "0.32957405",
        "pitch_frames": "[[...]]",
        "intensity": "-52.960995579585294",
        "intensity_frames": "[[...]]",
        "volume": "[...]",
        "avg_volume": "0.19482724",
        "loudness": "0.24210354685783386",
        "beats": "[  ...]",
        "segments_boundaries": "[...]",
        "segments_labels": "[...]",
        "frequency": "230.88860214871994",
        "key": "A#3",
        "frequencyFramesPath": "C:\\Users\\...\\polymath\\content\\db\\frequencyFrames\\song_name_ID_FrequencyFrames.txt"
      }
    }

Install

git clone https://github.com/hmbemba/polymath.git

cd polymath

pip install -r requirements.txt

Usage

• These are the current core features with more to come

Add Song

• Add a local song to the database

python cli.py addSong "path/to/song.mp3" 

splitstems

• The "-ss" boolean flag will split the stems right after

python cli.py addSong "path/too/song.mp3"  -ss

Add Video

• Download a single song from youtube and add it to the database

• Note : This is meant for a single video only not a playlist

python cli.py addVideo "https://youtu.be/dQw4w9WgXcQ"

Split Song

python cli.py split "path/to/song.mp3" "output/dir" 

Credits

• This project is a fork of the original Polymath project.
• The original project was developed by samim, and can be found at https://github.com/samim23/polymath