Generative Text to Audio – Sound Effects Using Audiogen
Practical AI through Prototypes
Music
Make a clip
CC BY 3.0 · Practical AI through Prototypes
Up to 60 seconds, at most the picture size of the copy it is cut from. The clip carries the title, the credit, the licence and a link back here. It is made in this browser, and nothing is sent to this site.
Or send the moment itself: https://invisiblenetwork.ai/watch/archive/commons/152523768?play=1&t=0
Parts
About
In this video we'll use the audiogen functionality of Audiocraft to generate sound effects based on descriptions of sounds we want to hear, provided via the command line.
Starting code: https://github.com/heaversm/audiocraft-tutorials/tree/tutorials/01-audiogen-cmdline Ending code: https://github.com/heaversm/audiocraft-tutorials/tree/tutorials/02-audiogen-params
About Me: I am a Staff Design Technologist on Mozilla's Innovation Team. All opinions and explorations are my own. Learn more about Mozilla Innovation at future.mozilla.org
0:00 - Introduction and Goals 0:13 - Stable Audio demo 0:33 - Audiocraft Github Repo 0:50 - Recap of last tutorial 1:30 - Intro to Audiogen 1:50 - Basic audio generation via command line 3:09 - Listening to our generated audio 3:50 - Set up script to accept command line parameters 6:24 - Testing generation via description parameters
6:55 - Next episode: Setting up a website form for audio generation
Human-made No record of this work says a model made it. What this means.