Resources: Difference between revisions

From Stop Synthetic Filth! wiki
Jump to navigation Jump to search
(→‎Commercial: + == Consumer protection == + https://www.dessa.com/ by w:Square, Inc. develops e.g. 'Fake voice detection' at github.com - "Using temporal convolution to detect audio deepfakes")
Line 22: Line 22:
* https://lalaland.ai/ - "''Artificial human models - Photorealistic yet nonexisting human models for eCommerce apparel webshops''"
* https://lalaland.ai/ - "''Artificial human models - Photorealistic yet nonexisting human models for eCommerce apparel webshops''"
* https://gradient.photo/ - "''The next big thing in the world of Mobile Photo Editing''" Android and iOS apps
* https://gradient.photo/ - "''The next big thing in the world of Mobile Photo Editing''" Android and iOS apps
== Consumer protection ==
* https://www.dessa.com/ by [[w:Square, Inc.]] develops e.g. [https://github.com/dessa-oss/fake-voice-detection ''''''Fake voice detection'''''' at github.com] - "''Using temporal convolution to detect audio deepfakes''"


= Syntheses =
= Syntheses =

Revision as of 19:00, 7 December 2020

You can find SSF! wiki resources. See Mediatheque for media. See our glossary for definitions.

Explore work by academia

Commercial

[1st seen in 1]

Visual synthetic human-like fakes

Consumer protection

Syntheses

Digital look-alikes

Links about digital look-alikes

Wikipedia links about digital look-alikes

Digital sound-alikes

Wikipedia links about digital sound-alikes

Other contenders to create digital sound-alikes

As of 2019, these speech synthesis in most use scenarios does not yet fool a human because the results contain tell tale signs that give it away as a speech synthesizer.


GANs

Links about StyleGAN


Text syntheses


Protective measures

Voice changers

List of voice changers

Multi-platform

  • pyvoicechanger is a real time microphone voice changer app for Python 3.6+
  • SOX is available for all OS and is usually pre-installed on Linux. Sox is a general purpose audio software, so configuration work is needed to enable the voice changers.

Windows


First seen in


Cite error: <ref> tags exist for a group named "1st seen in", but no corresponding <references group="1st seen in"/> tag was found