| Version 2 (modified by , 4 years ago) ( diff ) |
|---|
Adversarial Machine Learning for Voice Controllable Systems
Project Objective
Adversarial samples are intentionally designed to mislead trained machine learning models into making wrong predictions. This project will leverage adversarial machine learning techniques to attack voice controllable systems.
Reading Material
Week 1 Activites
- Get ORBIT/COSMOS account and familiarize oneself with the testbed procedures
Attachments (12)
- Imperceptible, Robust, and Targeted Adversarial Examples for Automatic Speech Recognition.pdf (368.2 KB ) - added by 4 years ago.
- Hidden voice commands.pdf (743.3 KB ) - added by 4 years ago.
- Audio Adversarial Examples Targeted Attacks on Speech-to-Text.pdf (587.5 KB ) - added by 4 years ago.
- Commandersong A systematic approach for practical adversarial voice recognition.pdf (824.7 KB ) - added by 4 years ago.
- Practical Adversarial Attacks Against Speaker Recognition Systems.pdf (1.9 MB ) - added by 4 years ago.
- hidden voice command code readme.docx (15.8 KB ) - added by 4 years ago.
- Weekly plan for adversarial machine learning against voice assistant systems.docx (7.5 KB ) - added by 4 years ago.
- lec5_neural network basic.pdf (2.0 MB ) - added by 4 years ago.
- X-VECTORS- ROBUST DNN EMBEDDINGS FOR SPEAKER RECOGNITION.pdf (189.6 KB ) - added by 4 years ago.
- Phoneme Recognition Using Time-Delay Neural Networks .pdf (1.2 MB ) - added by 4 years ago.
- Weekly plan for adversarial machine learning against voice assistant systems.2.docx (13.0 KB ) - added by 4 years ago.
- Probabilistic Linear Discriminant Analysis for Inferences About Identity.pdf (605.3 KB ) - added by 4 years ago.
Note:
See TracWiki
for help on using the wiki.
