Skip to content

Experiment with "one-shot learning" techniques to recognize a voice signature

License

Notifications You must be signed in to change notification settings

zdmc23/oneshot-audio

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

oneshot-audio

license

Experiment with "one-shot learning" techniques to recognize a voice signature. This was my submission and demo for LVTech hack-a-thon 2018

Alt text

One-Shot Learning?

"People can learn a new concept from just one or a few examples, making meaningful generalizations that go far beyond the observed data. Replicating this ability in machines has been challenging, since standard learning algorithms require tens, hundreds, or thousands of examples before reaching a high level of classification performance" [1]. (With Deep Learning, the dataset requirement has grown to 100K+).

  1. Lake, Lee, et. al. One-shot learning of generative speech concepts. 2014. (https://groups.csail.mit.edu/sls/publications/2014/lake-cogsci14.pdf).

Siamese Network

Alt text

Releases

No releases published

Packages

No packages published

Languages