Completed
FEATURES • MFCC-Mel-Frequency Cepstral Coefficients • Statistical features are used mean and variance
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Gyrophone - Eavesdropping Using a Gyroscope
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 INTRODUCTION - YAN MICHALEVSKY
- 3 MICROPHONE ACCESS
- 4 GYROSCOPE ACCESS BY A BROWSER JAVASCRIPT CODE EXAMPLE
- 5 MEMS GYROSCOPES
- 6 GYROSCOPES ARE SUSCEPTIBLE TO SOUND
- 7 GYROSCOPES ARE (LOUSY, BUT STILL) MICROPHONES
- 8 SOFTWARE LIMITATION OF THE SAMPLING RATE
- 9 SAMPLING FREQUENCY LIMITS
- 10 THE EFFECTS OF LOW SAMPLING FREQUENCY SPEECH SAMPLED AT 8,000HZ
- 11 WE CAN SENSE HIGH FREQUENCY SIGNALS DUE TO ALIASING
- 12 EXPERIMENTAL SETUP
- 13 SPEECH ANALYSIS USING A SINGLE GYROSCOPE
- 14 PREPROCESSING • All samples are converted to audio files in WAV format
- 15 FEATURES • MFCC-Mel-Frequency Cepstral Coefficients • Statistical features are used mean and variance
- 16 CLASSIFIERS
- 17 DYNAMIC TIME WARPING
- 18 GENDER IDENTIFICATION
- 19 WE CAN SUCCESSFULLY IDENTIFY GENDER
- 20 SPEAKER IDENTIFICATION
- 21 A GOOD CHANCE TO IDENTIFY THE SPEAKER
- 22 ISOLATED WORDS RECOGNITION SPEAKER INDEPENDENT
- 23 HOW CAN WE LEVERAGE EAVESDROPPING SIMULTANEOUSLY ON TWO DEVICES?
- 24 SIMILAR TO TIME-INTERLEAVED ADC's
- 25 NON-UNIFORM RECONSTRUCTION REQUIRES KNOWING PRECISE TIME-SKEWS
- 26 PRACTICAL COMPROMISE Interleaving samples from multiple devices
- 27 EVALUATION Tested for the case of speaker dependent word recognition
- 28 FURTHER ATTACKS
- 29 SOURCE SEPARATION
- 30 AMBIENT SOUND RECOGNITION
- 31 SOFTWARE DEFENSES
- 32 HARDWARE DEFENSES
- 33 CONCLUSION
