Skip to content
Music identification using spectrogram based peak-pair hashing
Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.gitignore
Pipfile
Pipfile.lock
README.md
audio.py
audio_convert.py
config.py
database.py
fingerprint.py
lookup.py
save_songs.py
song.py
utils.py

README.md

soundz

Audio fingerprinting and identification in Python. See the dev blog for more details.

Build

This is a Python 3 application that uses Pipenv.

  • Install the dependencies
> pipenv install
  • Activate the shell
> pipenv shell

Saving Songs

A PostgreSQL database is assumed to be running on port 5432 with the username and password being "soundz". The database url can be changed in database.py.

Songs can be added and used for later identication with

> python save_songs.py FILE|DIRECTORY

where FILE is the relative or absolute path of an audio file and DIRECTORY is the relative or absolute path of an directory containing audio files. The only supported audio formats at the moment are mp3 and mp4.

Identification

Audio identification is done using the microphone. The command

> python lookup.py

can be used to record and identify 10 seconds of audio. The amount of time can be changed in lookup.py

You can’t perform that action at this time.