fat84/audioprint

This repository was created in order to evaluate an fingerprinting algorithm with some degraded audio samples.

AudioFingerprintTest

---

This repository was created in order to evaluate an fingerprinting algorithm with some degraded audio samples, see the parametrization process here ArtículoETCM2017.pdf

Videos

Multimedia content that was uploaded to the server

AudioSamples

Some degraded audio samples with various size and #SNR (file_(size)SNR(#).wav)

sendfile.py

Python script for connecting to server and loading an audio sample in order to retrieve a link of additional information about the audio sample. By default the script uses input.wav, but it could be changed by another file with the following extension "wav, mp4, mp3, 3gp"

file='input.wav'	

input.wav

This file will be uploaded to the server

Quickstart

$ git clone https://github.com/JoseLMedinaC/AudioFingerprintTest.git ./AudioFingerprintTest
$ cd AudioFingerprintTest
$ python sendfile.py

Results

This plot is the result of computing various audios samples varying SNR and modularity.

Precisión=>Accuracy

FP========>False Positive Rate