Freedom251 › AI Deepfake Detection Tools
DeepFake-O-Meter
What DeepFake-O-Meter is
DeepFake-O-Meter is a free web platform for examining whether images, videos, or audio may have been generated by AI. It brings together 37 forensic research models, with catalog entries covering all three media types. Users can select algorithms using listed measures such as accuracy, running time, and year developed. Each selected algorithm returns a percentage indicating the likelihood that the submitted content was AI-generated.
The platform is attributed to the University at Buffalo and its UB Media Forensics Lab; the site names Siwei Lyu as project director. Its intended audience includes social media users, journalists, and law enforcement. DeepFake-O-Meter is described as open source, with algorithm source code publicly accessible. Before uploading media, users are asked whether they want to share it with researchers; shared submissions can help train detection algorithms. The platform presents results from multiple methods and leaves users to decide whether content is real. One catalog note says WAV2LIP-STA performance may suffer under heavy compression. The home page reports that the service was used on the AFP fact check website to detect a deepfake video.
Who it suits
DeepFake-O-Meter suits social media users, journalists, and law enforcement who want to examine images, video, or audio with forensic research models. It may also suit people interested in publicly accessible algorithm source code and comparing outputs from multiple detection methods.
What it does well
- Analyzes images, videos, and audio.
- Lists 37 integrated forensic research models.
- Users can select algorithms by accuracy, running time, and development year.
- Each selected algorithm returns a likelihood percentage.
- Algorithm source code is publicly accessible.
- Free web access.
What to know first
- WAV2LIP-STA may perform worse under heavy compression.
- Results leave authenticity judgments to the user.
Getting started
- Visit the DeepFake-O-Meter website.
- Create an account with an email, username, and password.
- Choose detection algorithms using the listed model metrics.
- Before uploading, decide whether to share the media with researchers.
- Review each selected algorithm's likelihood percentage.
Verdict
Choose DeepFake-O-Meter if you want free web access to multiple research models for examining images, video, or audio. Look elsewhere if you need the platform itself to make a definitive authenticity judgment; its outputs are analyses for users to interpret.
Questions
Is DeepFake-O-Meter free?
Yes. Its pricing model is free, and a free plan is listed.
Which media types can it analyze?
The platform analyzes images, videos, and audio.
How many detection models are listed?
The models page lists 37 integrated forensic research models.
Is DeepFake-O-Meter open source?
It is described as open source, with algorithm source code publicly accessible.
Who develops DeepFake-O-Meter?
The site attributes it to the University at Buffalo and the UB Media Forensics Lab, and lists Siwei Lyu as project director.
Does the platform decide whether media is authentic?
It provides analysis from multiple methods and leaves users to decide whether content is real.
Facts from DeepFake-O-Meter's own pages
- Account access
- The site provides account sign-up and login pages, and its login page offers a guest option to return to the home page. · zinc.cse.buffalo.edu, 7 Oct 2026
- Current use
- The home page reports that Deepfake-o-Meter was used on the AFP fact check website to detect a deepfake video. · zinc.cse.buffalo.edu, 7 Oct 2026
- Detection models
- The models page lists 37 integrated forensic research models. · zinc.cse.buffalo.edu, 7 Oct 2026
- Headquarters
- Buffalo, New York, United States · zinc.cse.buffalo.edu, 28 Sept 2026
- Intended users
- The University at Buffalo describes the platform as serving users including social media users, journalists, and law enforcement. · buffalo.edu, 7 Oct 2026
- Interpretation caveat
- The platform provides analysis from multiple methods and does not make strong claims about whether uploaded content is authentic. · buffalo.edu, 7 Oct 2026
- Maker
- The site attributes DeepFake-O-Meter to the University at Buffalo and the UB Media Forensics Lab. · zinc.cse.buffalo.edu, 7 Oct 2026
- Media types
- The platform analyzes images, videos, and audio. · buffalo.edu, 7 Oct 2026
- Model categories
- The model catalog includes methods labeled for image, video, and audio detection. · zinc.cse.buffalo.edu, 7 Oct 2026
- Model limitations
- The model catalog notes that WAV2LIP-STA performance may suffer under heavy compression. · zinc.cse.buffalo.edu, 7 Oct 2026
- Model selection
- Users can select detection algorithms using listed metrics that include accuracy, running time, and year developed. · buffalo.edu, 7 Oct 2026
- Multiple analyses
- The University at Buffalo says users can run multiple detection algorithms, each returning a percentage likelihood that content was AI generated. · buffalo.edu, 7 Oct 2026
- Open source
- The platform is described as open source, with algorithm source code publicly accessible. · buffalo.edu, 7 Oct 2026
- Project team
- The site identifies the UB Media Forensics Lab as the developer and lists Siwei Lyu as project director. · zinc.cse.buffalo.edu, 7 Oct 2026
- Purpose
- DeepFake-o-Meter aggregates results from research AI media detection models to support media authenticity assessment. · zinc.cse.buffalo.edu, 7 Oct 2026
- Research models
- The detection models page lists 37 integrated forensic research models. · zinc.cse.buffalo.edu, 7 Oct 2026
- Result interpretation
- The University at Buffalo says the platform provides analysis from a broad range of methods and leaves users to decide whether content is real. · buffalo.edu, 7 Oct 2026
- Results
- Each selected algorithm returns a percentage representing the likelihood that content was AI-generated. · buffalo.edu, 7 Oct 2026
- Submission sharing
- Before upload, users are asked whether they want to share the media with researchers; the team says shared submissions can help train detection algorithms. · buffalo.edu, 7 Oct 2026
- Support contact
- The contact page lists [email protected] for contacting the lab. · zinc.cse.buffalo.edu, 7 Oct 2026
- User audience
- The maker describes the platform as intended to make deepfake analysis available to the public, including social media users, journalists, and law enforcement. · buffalo.edu, 7 Oct 2026
Where DeepFake-O-Meter stands on Freedom251
- AI Deepfake Detection Tools#9 of 20