The Deepfakes Podcast
By Ishan Sharma (Deepsync Technologies)
Deepsync usually requires at least 10 hours of audio to generate a high-quality replica. This voice was produced from just 2 hours of the host's old recordings but clearly shows the power of Deepsync in reducing production complexity.
The Deepfakes PodcastSep 11, 2021
Charles Bukowski's poetry in my voice
With only 2 hours of my AI voice, I can now create content on the go, from literature to recipes to nutrition guides (more coming soon). Here I give a try to a poem by a favorite writer - Charles Bukowski 👇
For High-quality voices or creating your own custom voice, check out deepsync.co ↗️
This is a boon to brands and podcasting houses, helping 10x the production while still retaining the quality processes!
Note: All rights belong to the writer and the publisher. This is intended as research.
#ai #research #charlesbukowski #syntheticmedia
The Poem:
16-bit Intel 8088 chip
with an Apple Macintosh
you can't run Radio Shack programs
in its disc drive.
nor can a Commodore 64
drive read a file
you have created on an
IBM Personal Computer.
both Kaypro and Osborne computers use
the CP/M operating system
but can't read each other's
handwriting
for they format (write
on) discs in different
ways.
the Tandy 2000 runs MS-DOS but
can't use most programs produced for
the IBM Personal Computer
unless certain
bits and bytes are
altered
but the wind still blows over
Savannah
and in the Spring
the turkey buzzard struts and
flounces before his
hens.
Short-podcasts meet AI
Today, 68 million Americans listen to podcasts on a weekly basis. In our journey to speak about the potential good of synthetic media, here we present an audio version of our blog, read in the host's AI voice. Short Podcasts or Micro casts represent the bite-sized future of audio content. They are growing quickly and offer opportunities for both individuals and brands.
Complete article can be found here: https://guides.deepsync.co/en/articles/5525604-short-podcasts-the-bite-sized-future-of-audio-content
--------------
Disclaimer: The host's AI voice was created using only 2 hours of his previously recorded audio by Deepsync, a pioneer of voice cloning technology. Deepsync usually requires upwards of 10 hours of audio to create a perfect AI voice. Learn more at www.deepsync.co
Welcome to The Deepfakes Podcast
Welcome to the world's first of its kind podcast talking about the good and the bad of Deepfakes and produced entirely using the synthetic audio of the host's voice. This show is about educating everyone about what synthetic audio is, how it works, how we can combat Deepfakes, and the future of this technology to empower podcast production. Produced by Deepsync, The technology requires upwards of 10 hours of audio to produce a perfect replica of one's voice but for the sake of simplicity, this podcast was produced using only a couple of hours of the host's original audio files and then creating digital audio with it.
You can check out deepsync.co to learn more about this groundbreaking technology.