WaveFake: A Data Set to Facilitate Audio DeepFake Detection

Related tags

Data AnalysisWaveFake
Overview

WaveFake: A Data Set to Facilitate Audio DeepFake Detection

logo

This is the code repository for our NeurIPS 2021 (Track on Datasets and Benchmarks) paper WaveFake.

Deep generative modeling has the potential to cause significant harm to society. Recognizing this threat, a magnitude of research into detecting so-called "Deepfakes" has emerged. This research most often focuses on the image domain, while studies exploring generated audio signals have - so far - been neglected. In this paper, we aim to narrow this gap. We present a novel data set, for which we collected ten sample sets from six different network architectures, spanning two languages. We analyze the frequency statistics comprehensively, discovering subtle differences between the architectures, specifically among the higher frequencies. Additionally, to facilitate further development of detection methods, we implemented three different classifiers adopted from the signal processing community to give practitioners a baseline to compare against. In a first evaluation, we already discovered significant trade-offs between the different approaches. Neural network-based approaches performed better on average, but more traditional models proved to be more robust.

Dataset & Pre-trained Models

You can find our dataset on zenodo and we also provide pre-trained models.

Setup

You can install all needed dependencies by running:

pip install -r requirements.txt

RawNet2 Model

For consistency, we use the RawNet2 model provided by the ASVSpoof 2021 challenge. Please download the model specifications here and place it under dfadetect/models as raw_net2.py.

Statistics & Plots

To recreate the plots/statistics of the paper, use:

python statistics.py -h

usage: statistics.py [-h] [--amount AMOUNT] [--no-stats] [DATASETS ...]

positional arguments:
  DATASETS              Path to datasets. The first entry is assumed to be the referrence one. Specified as follows <path,name>

optional arguments:
  -h, --help            show this help message and exit
  --amount AMOUNT, -a AMOUNT
                        Amount of files to concider.
  --no-stats, -s        Do not compute stats, only plots.

Example

python statistics.py /path/to/reference/data,ReferenceDataName /path/to/generated/data,GeneratedDataName -a 10000

Training models

You can use the training script as follows:

python train_models.py -h

usage: train_models.py [-h] [--amount AMOUNT] [--clusters CLUSTERS] [--batch_size BATCH_SIZE] [--epochs EPOCHS] [--retraining RETRAINING] [--ckpt CKPT] [--use_em] [--raw_net] [--cuda] [--lfcc] [--debug] [--verbose] REAL FAKE

positional arguments:
  REAL                  Directory containing real data.
  FAKE                  Directory containing fake data.

optional arguments:
  -h, --help            show this help message and exit
  --amount AMOUNT, -a AMOUNT
                        Amount of files to load from each directory (default: None - all).
  --clusters CLUSTERS, -k CLUSTERS
                        The amount of clusters to learn (default: 128).
  --batch_size BATCH_SIZE, -b BATCH_SIZE
                        Batch size (default: 8).
  --epochs EPOCHS, -e EPOCHS
                        Epochs (default: 5).
  --retraining RETRAINING, -r RETRAINING
                        Retraining tries (default: 10).
  --ckpt CKPT           Checkpoint directory (default: trained_models).
  --use_em              Use EM version?
  --raw_net             Train raw net version?
  --cuda, -c            Use cuda?
  --lfcc, -l            Use LFCC instead of MFCC?
  --debug, -d           Only use minimal amount of files?
  --verbose, -v         Display debug information?

Example

To train all EM-GMMs use:

python train_models.py /data/LJSpeech-1.1/wavs /data/generated_audio -k 128 -v --use_em --epochs 100

Evaluation

For evaluation you can use the evaluate_models script:

python evaluate_models.p -h

usage: evaluate_models.py [-h] [--output OUTPUT] [--clusters CLUSTERS] [--amount AMOUNT] [--raw_net] [--debug] [--cuda] REAL FAKE MODELS

positional arguments:
  REAL                  Directory containing real data.
  FAKE                  Directory containing fake data.
  MODELS                Directory containing model checkpoints.

optional arguments:
  -h, --help            show this help message and exit
  --output OUTPUT, -o OUTPUT
                        Output file name.
  --clusters CLUSTERS, -k CLUSTERS
                        The amount of clusters to learn (default: 128).
  --amount AMOUNT, -a AMOUNT
                        Amount of files to load from each directory (default: None - all).
  --raw_net, -r         RawNet models?
  --debug, -d           Only use minimal amount of files?
  --cuda, -c            Use cuda?

Example

python evaluate_models.py /data/LJSpeech-1.1/wavs /data/generated_audio trained_models/lfcc/em

Make sure to move the out-of-distribution models to a seperate directory first!

Attribution

We provide a script to attribute the GMM models:

python attribute.py -h

usage: attribute.py [-h] [--clusters CLUSTERS] [--steps STEPS] [--blur] FILE REAL_MODEL FAKE_MODEL

positional arguments:
  FILE                  Audio sample to attribute.
  REAL_MODEL            Real model to attribute.
  FAKE_MODEL            Fake Model to attribute.

optional arguments:
  -h, --help            show this help message and exit
  --clusters CLUSTERS, -k CLUSTERS
                        The amount of clusters to learn (default: 128).
  --steps STEPS, -m STEPS
                        Amount of steps for integrated gradients.
  --blur, -b            Compute BlurIG instead.

Example

python attribute.py /data/LJSpeech-1.1/wavs/LJ008-0217.wav path/to/real/model.pth path/to/fake/model.pth

BibTeX

When you cite our work feel free to use the following bibtex entry:

@inproceedings{
  frank2021wavefake,
  title={{WaveFake: A Data Set to Facilitate Audio Deepfake Detection}},
  author={Joel Frank and Lea Sch{\"o}nherr},
  booktitle={Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track},
  year={2021},
}
Owner
Chair for Sys­tems Se­cu­ri­ty
Chair for Sys­tems Se­cu­ri­ty
Program that predicts the NBA mvp based on data from previous years.

NBA MVP Predictor A machine learning model using RandomForest Regression that predicts NBA MVP's using player data. Explore the docs » View Demo · Rep

Muhammad Rabee 1 Jan 21, 2022
fds is a tool for Data Scientists made by DAGsHub to version control data and code at once.

Fast Data Science, AKA fds, is a CLI for Data Scientists to version control data and code at once, by conveniently wrapping git and dvc

DAGsHub 359 Dec 22, 2022
Data science/Analysis Health Care Portfolio

Health-Care-DS-Projects Data Science/Analysis Health Care Portfolio Consists Of 3 Projects: Mexico Covid-19 project, analyze the patient medical histo

Mohamed Abd El-Mohsen 1 Feb 13, 2022
A real data analysis and modeling project - restaurant inspections

A real data analysis and modeling project - restaurant inspections Jafar Pourbemany 9/27/2021 This project represents data analysis and modeling of re

Jafar Pourbemany 2 Aug 21, 2022
A Python adaption of Augur to prioritize cell types in perturbation analysis.

A Python adaption of Augur to prioritize cell types in perturbation analysis.

Theis Lab 2 Mar 29, 2022
Amundsen is a metadata driven application for improving the productivity of data analysts, data scientists and engineers when interacting with data.

Amundsen is a metadata driven application for improving the productivity of data analysts, data scientists and engineers when interacting with data.

Amundsen 3.7k Jan 03, 2023
Open-source Laplacian Eigenmaps for dimensionality reduction of large data in python.

Fast Laplacian Eigenmaps in python Open-source Laplacian Eigenmaps for dimensionality reduction of large data in python. Comes with an wrapper for NMS

17 Jul 09, 2022
Python library for creating data pipelines with chain functional programming

PyFunctional Features PyFunctional makes creating data pipelines easy by using chained functional operators. Here are a few examples of what it can do

Pedro Rodriguez 2.1k Jan 05, 2023
BErt-like Neurophysiological Data Representation

BENDR BErt-like Neurophysiological Data Representation This repository contains the source code for reproducing, or extending the BERT-like self-super

114 Dec 23, 2022
Spaghetti: an open-source Python library for the analysis of network-based spatial data

pysal/spaghetti SPAtial GrapHs: nETworks, Topology, & Inference Spaghetti is an open-source Python library for the analysis of network-based spatial d

Python Spatial Analysis Library 203 Jan 03, 2023
ETL pipeline on movie data using Python and postgreSQL

Movies-ETL ETL pipeline on movie data using Python and postgreSQL Overview This project consisted on a automated Extraction, Transformation and Load p

Juan Nicolas Serrano 0 Jul 07, 2021
InDels analysis of CRISPR lines by NGS amplicon sequencing technology for a multicopy gene family.

CRISPRanalysis InDels analysis of CRISPR lines by NGS amplicon sequencing technology for a multicopy gene family. In this work, we present a workflow

2 Jan 31, 2022
SNV calling pipeline developed explicitly to process individual or trio vcf files obtained from Illumina based pipeline (grch37/grch38).

SNV Pipeline SNV calling pipeline developed explicitly to process individual or trio vcf files obtained from Illumina based pipeline (grch37/grch38).

East Genomics 1 Nov 02, 2021
ForecastGA is a Python tool to forecast Google Analytics data using several popular time series models.

ForecastGA is a tool that combines a couple of popular libraries, Atspy and googleanalytics, with a few enhancements.

JR Oakes 36 Jan 03, 2023
follow-analyzer helps GitHub users analyze their following and followers relationship

follow-analyzer follow-analyzer helps GitHub users analyze their following and followers relationship by providing a report in html format which conta

Yin-Chiuan Chen 2 May 02, 2022
Containerized Demo of Apache Spark MLlib on a Data Lakehouse (2022)

Spark-DeltaLake-Demo Reliable, Scalable Machine Learning (2022) This project was completed in an attempt to become better acquainted with the latest b

8 Mar 21, 2022
PandaPy has the speed of NumPy and the usability of Pandas 10x to 50x faster (by @firmai)

PandaPy "I came across PandaPy last week and have already used it in my current project. It is a fascinating Python library with a lot of potential to

Derek Snow 527 Jan 02, 2023
Randomisation-based inference in Python based on data resampling and permutation.

Randomisation-based inference in Python based on data resampling and permutation.

67 Dec 27, 2022
Feature Detection Based Template Matching

Feature Detection Based Template Matching The classification of the photos was made using the OpenCv template Matching method. Installation Use the pa

Muhammet Erem 2 Nov 18, 2021
talkbox is a scikit for signal/speech processing, to extend scipy capabilities in that domain.

talkbox is a scikit for signal/speech processing, to extend scipy capabilities in that domain.

David Cournapeau 76 Nov 30, 2022