10000 GitHub - thekevinscott/laugh-tracks: Models and examples built with TensorFlow
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content
This repository was archived by the owner on Aug 11, 2019. It is now read-only.

thekevinscott/laugh-tracks

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Laughter from a Can

Spam

This is a project to teach a machine to provide canned laugh tracks to normal, everyday life. The goal is to use Machine Learning to listen to an audio stream and identify when the optimal time is to insert some laughter.

Background & Tools

The project is separated into multiple phases, aimed at building increasing levels of understanding.

Phase One

First, we need to generate a corpus of laughter.

VGGish is a pretrained model for identifying audio.

Getting Started

There is a Dockerfile to make it easy to spin up. Build and run with:

docker build -t laugh-tracks .
nvidia-docker run -it --rm -v $(pwd):/notebooks/ -p 8889:8888  --name laugh-tracks laugh-tracks

or in fish shell:

docker build -t laugh-tracks .
4FA2

nvidia-docker run -it --rm -v (pwd):/notebooks/ -p 8889:8888  --name laugh-tracks laugh-tracks

About

Models and examples built with TensorFlow

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Jupyter Notebook 99.6%
  • Other 0.4%
0