New Whitepaper: Securing the Future of Voice with OmniSpeech AI Detectâ„¢
- OmniSpeech Team
- Sep 17
- 3 min read
Updated: Oct 1
New Gen 2 AI Detectâ„¢ Algorithm is Top Industry Performer in Detecting Deepfake Audio From Unseen Data Sets
Today we’re releasing a new whitepaper, Securing the Future of Voice: AI Deepfake Detection with OmniSpeech AI Detect™, authored by Dr. Carol Espy-Wilson (PhD), Yashish Maduwantha (PhD), and David Przygoda (MBA/MS) for OmniSpeech LLC. It lays out the deepfake voice threat, the science behind our Gen 2 detection engine, and the results that matter for real-world deployments.
Why this matters
Voice cloning is now cheap, fast, and convincing. With seconds of audio, attackers can spoof executives, trigger fraudulent wire transfers, or seed misinformation. OmniSpeech AI Detect™ brings speech-science-grounded deep learning and real-time signal processing to identify synthetic voices—live or offline—across languages, accents, and most importantly, from unseen datasets.
Key results (highlights)
ASVSpoof 5 (test split) – Equal Error Rate (EER%)
OmniSpeech AI Detectâ„¢ Gen 1: 8.43 - Best EER score for the ASVSpoof 5 test split per the Hugging Face Speech Deepfake Arena.
Selected industry/open-source baselines:
Whispeak 9.92
Syntra 15.96
Resemble Detect 16.29,
Wav2vec2 AASIST 16.24
XLSR+SLS 18.76
AASIST 35.53.
Cross-corpus generalization (unseen data)
Trained on 2019–2025 deepfake systems and evaluated on ASVSpoof 5, SONAR, DFADD, EmoFake:
Layering proprietary speech science on top of traditional classification models, OmniSpeech AI Detectâ„¢ Gen 2 saw an Avg. EER: 4.62% | Avg. Acc: 93.33% - This is a new standard for models in generalizing across unseen datasets, speech generators, and acoustic conditions.
Gen 1 Avg. EER: 45.40% | Avg. Acc: 57.06%
Use cases
Enterprise & Government: Reinforce voice biometrics, stop call-center fraud in-stream, and validate emergency calls.
Social & Content Platforms: Moderate synthetic voice at upload/stream time; certify podcast and livestream authenticity.
Consumer: Protect voice assistants from replay/deepfake commands; warn users during AI-generated scam calls.
What’s inside the whitepaper
Threat landscape & impact. A concise overview of the acceleration in AI-generated voice abuse and its financial/social costs.
How our detector works. Deep learning models infused with decades of speech production research, trained on custom, diverse corpora of real and synthetic voices.
Built for reality. Robust to codecs, compression, and environmental noise, and capable of catching partial spoofs (human speech with synthetic inserts).
AI Detectâ„¢ API available now for select partners; broader platform availability to follow. Download the whitepaper, "Securing the Future of Voice: AI Deepfake Detection with OmniSpeech AI Detectâ„¢":
Availability and Demos
In addition to a new consumer/enterprise level app for Zoom coming soon, OmniSpeech’s AI Detect™ deepfake detection is now available via API for licensing to partners across industries. Manufacturers, developers, and security teams can explore the solution’s capabilities and integrate it into their systems to enhance security and build trust while securing new revenue streams.
To schedule a demo or learn more, contact partnerships@omni-speech.com.
© 2025 OmniSpeech LLC. All rights reserved. OmniSpeech® and OmniSpeech AI Detect™ are trademarks or registered trademarks of OmniSpeech LLC. Performance may vary by environment and configuration; see the whitepaper’s legal disclaimers for details.
###
About OmniSpeech - OmniSpeech is a pioneer in AI voice technology, dedicated to enhancing voice experiences on any app or device. From noise suppression to advanced speech analysis, OmniSpeech’s solutions are transforming the way businesses, devices, and individuals interact with voice technology. OmniSpeech is a graduate of the Venture Accelerator program at the University of Maryland and the Advanced Technology Development Center (ATDC) Accelerate program. The company’s innovations have earned prestigious industry awards and licensing agreements with Fortune 500 companies. For more information, visit: https://omni-speech.com