Sorry, you need to enable JavaScript to visit this website.

Electric Network Frequency(ENF) is a recently developed technique for the authentication of audio signals. ENF gets embedded in audio signals due to electromagnetic interferences from power lines and hence can be used as a measure to find the geographical location and time of recording. Given an audio signal, this paper presents a technique of finding the power grid the audio belongs to. Towards this, we first extract the ENF sinusoid using a very narrow bandwidth filter centered around the nominal frequency. The filter is designed using a frequency response masking approach.

This report presents the results to the challenge of
”Exploring Power Signatures for Location Forensics of Media
Recordings” as apart of Signal Processing Cup 2016 by IEEE
Signal Processing Society. Here we examine different frequency
estimation and classification techniques to provide accurate ENF
estimates and classify these signals into the corresponding grid
of recording. In this report we propose methods of efficient
extraction of ENF signal using quadratic interpolation and
frequency tracking.The SVM and GMM classifiers used provided

Electrical network frequency (ENF) has been used as evidence for location forensic. To determine location, we need accurate ENF information from noisy media files, select features of the ENF signal and the classify it based on previous knowledge of different grids.

Deep neural networks (DNN) are a powerful tool for many large vocabulary continuous speech recognition (LVCSR) tasks. Training a very deep network is a challenging problem and pre-training techniques are needed in order to achieve the best results. In this paper, we propose a new type of network architecture, Linear Augmented Deep Neural Network (LA-DNN). This type of network augments each non-linear layer with a linear connection from layer input to layer output.

Slides from the presentation held at ICASSP 2016 for the paper: Recurrent neural networks for polyphonic sound event detection in real life recordings

This paper presents a new method for solving linear inverse problems where the observations are corrupted with a mixed Poisson-Gaussian noise.

In the state of the art, a single frame of DFT transform is commonly used as a basis for building amplitude spectral envelopes.
Multiple Frame Analysis (MFA) has already been suggested for envelope estimation, but often with excessive complexity.
In this paper, two MFA-based methods are presented: one simplifying an existing Least Square (LS) solution, and another one based on a simple linear interpolation.