# Aurigin.ai > Aurigin.ai is a Swiss deep-tech security company specializing in real-time audio deepfake detection, anti-spoofing, and liveness detection for voice biometric systems. We help voice biometric companies, contact centers, enterprises, and forensic investigators detect AI-generated, cloned, and synthetic voices with 98%+ accuracy, sub-50ms latency, across 80+ languages. ## Products - [AuriginAPI](https://aurigin.ai/aurigin-api): REST API, WebSocket streaming, and native SDKs (Python, Node.js, Java) for real-time voice deepfake detection and identity verification. 98%+ accuracy, sub-50ms latency, 80+ languages. Free tier available. - [AuriginGuard](https://aurigin.ai/aurigin-guard): Enterprise desktop application for real-time vishing protection during video conferencing. Detects AI-generated voices, verifies caller identity, and identifies social engineering patterns across Microsoft Teams, Zoom, Google Meet, and Webex. - [AuriginShield](https://aurigin.ai/aurigin-shield): Voice fraud prevention layer for contact centers. Deepfake detection, caller verification, bot filtering, trust scoring, and compliance audit trail. Integrates with Genesys, NICE CXone, Five9, RingCentral, and Cisco. - [AuriginInspect](https://aurigin.ai/aurigin-inspector): Forensic-grade deepfake detection platform for legal proceedings, law enforcement, and media investigations. 98.1% accuracy with court-ready reporting, segment localization, and explainable AI. ## Solutions / Use Cases - [Secure Video Conferencing](https://aurigin.ai/video-conferencing-security): Protect enterprise video calls from CEO fraud, vishing, and AI voice impersonation across Teams, Zoom, Google Meet, and Webex. - [Contact Center Security](https://aurigin.ai/contact-center-security): Prevent account takeover, payment redirect fraud, and MFA bypass via synthetic voice attacks on call centers. - [Forensics & Investigations](https://aurigin.ai/forensics-investigations): Court-ready audio and video deepfake analysis for forensic labs, law enforcement, newsrooms, and media verification. ## Pricing - [Pricing](https://aurigin.ai/pricing): Usage-based plans from Free tier to Enterprise with on-premise deployment, custom SLAs, and volume discounts. ## Resources - [Blog](https://aurigin.ai/blog): Technical articles and research on deepfake detection, voice biometrics security, anti-spoofing, and AI fraud trends. - [Latest News](https://aurigin.ai/latest-news): Press coverage and media mentions of Aurigin.ai. - [Industry Reports](https://aurigin.ai/industry-reports): White papers and research on deepfake fraud across banking, enterprise, and media. - [FAQ](https://aurigin.ai/faq): Frequently asked questions about Aurigin products, integration, accuracy, and data privacy. - [About](https://aurigin.ai/about): Company mission, vision, and team. ## Key Facts - Founded: 2024 - Headquarters: Switzerland (Swiss Made Software certified) - Recognition: #1 in Security at TOP 100 Swiss Startups 2025 Public Voting; Kickstart Innovation 2025 Cohort - Detection accuracy: 98%+ (Apollo model) - API latency: sub-50ms - Languages supported: 80+ - Deployment: Cloud API, on-premise (Docker/Kubernetes), desktop (Windows, macOS) - Integrations: Microsoft Teams, Zoom, Google Meet, Webex, Genesys, NICE CXone, Five9, RingCentral, Cisco - Team: Built by experts from ETH Zurich, Sennheiser, McKinsey, PwC, Lombard Odier ## Contact - Website: https://aurigin.ai - LinkedIn: https://www.linkedin.com/company/aurigin-ai/ - YouTube: https://www.youtube.com/@aurigin-ai ## Detailed Product Information ### AuriginAPI — Voice Deepfake Detection & Identity Verification API AuriginAPI is a developer-first REST API (with WebSocket streaming and native SDKs) that classifies any voice as real or synthetic in less than 3 lines of code. It is designed to slot into voice biometric pipelines, authentication systems, contact center platforms, and any application that processes voice. Key features: - Real-time streaming analysis with sub-50ms latency - 98%+ detection accuracy on benchmark datasets (TTS, voice conversion, replay, voice cloning) - Native SDKs for Python, Node.js, and Java - REST and WebSocket endpoints - On-premise deployment via Docker or Kubernetes (no audio leaves the environment) - Free tier, usage-based pricing, and enterprise plans ### AuriginGuard — Real-Time Vishing Protection for Enterprise AuriginGuard is a desktop application (Windows, macOS) that protects employees on video calls from AI-generated voice attacks. It runs silently alongside video conferencing platforms and alerts the employee in real time when a synthetic voice is detected. Key features: - Layer 1: Synthetic voice detection on every call - Layer 2: Caller identity verification against enrolled voiceprints - Layer 3: Social engineering pattern detection (urgency cues, payment requests) - Works across Microsoft Teams, Zoom, Google Meet, and Webex - On-device processing by default — privacy by design - Centralized enterprise admin console, audit logs ### AuriginShield — Voice Fraud Prevention for Contact Centers AuriginShield is the trust and fraud-prevention layer for modern contact centers. It screens every call for synthetic audio, verifies caller identity, filters bots, and outputs a calibrated trust score that routes risky calls to high-assurance flows. Key features: - Deepfake / synthetic voice detection on inbound calls - Caller identity verification and voiceprint matching - Bot and IVR-evasion filtering - Risk scoring and routing signals - Compliance-ready audit trail (PSD2, GLBA, SOC 2) - Integrations with Genesys, NICE CXone, Five9, RingCentral, and Cisco ### AuriginInspect — Forensic-Grade Audio Deepfake Detection AuriginInspect is the first forensic-grade deepfake detection platform purpose-built for legal proceedings, law enforcement, and investigative journalism. It produces explainable, court-admissible analysis of audio evidence. Key features: - Calibrated likelihood-ratio scoring for genuine vs synthetic audio - Segment localization — pinpoint which parts of a recording were manipulated - Explainable outputs with characterized uncertainty - Court-ready report generation - Chain-of-custody and evidence tracking - Secure air-gapped deployment for sensitive investigations ## Threats Addressed - Voice cloning and AI-generated synthetic speech (TTS and voice conversion) - Replay attacks against voice biometric systems - Vishing (voice phishing) and CEO fraud over video calls - Account takeover via contact centers - IT helpdesk impersonation and hiring fraud - Fabricated audio/video evidence in legal and media contexts ## Data Privacy & Compliance - Cloud API: audio is processed and immediately deleted; never used for training - On-premise: audio never leaves the customer environment - Enterprise-ready: SOC 2-aligned controls, Swiss data sovereignty, GDPR-compliant - No biometric identifiers stored unless explicitly configured by the customer ## About the Company Aurigin.ai was founded in 2024 in Switzerland by Luzi Sennhauser (CEO) and Simon Thomé (CPO). The company is backed by expertise from ETH Zurich, Sennheiser, McKinsey, PwC, and Lombard Odier. Aurigin.ai is certified as Swiss Made Software and was voted #1 in Security at TOP 100 Swiss Startups 2025 Public Voting. It is part of the Kickstart Innovation 2025 cohort. The team includes: - Luzi Sennhauser — Co-founder & CEO - Simon Thomé — Co-founder & CPO - Nicolas Wagner — Head of Engineering - Violeta Petrovic — DevOps Engineer - Kudret Esmer — Senior Data Scientist - Mirjam Kaeser — Business Operations Manager - Dániel Bányay — AI Research Engineer - Georgios Xypolitos — Security Engineer ## Industry Context Deepfake and synthetic voice fraud is growing rapidly: - 700%+ annual increase in deepfake fraud targeting banks - 442% surge in vishing attacks, fuelled by generative AI - $25M lost in a single deepfake video call impersonation (Arup, 2024) - $15.6B in account takeover losses in the US (2024) - 25% of bank fraud originates from contact centers - 8M+ deepfakes circulating online in 2025 Aurigin.ai addresses these threats with real-time, explainable, production-grade detection that integrates into existing voice workflows without disrupting the user experience. ## Latest Blog Posts ### Apollo V4: From Benchmarks to Everyday Trust - Date: 2026-02-19 - URL: https://aurigin.ai/blog/apollo-v4-from-benchmarks-to-everyday-trust - Summary: Apollo V4 extends industry-leading deepfake detection into the full complexity of real-world communication. Building on Apollo's position at the top of public benchmarks, V4 goes further, delivering robust protection across the compression pipelines of Microsoft Teams, Google Meet, and Zoom, through codec degradation and network variability, and across 40+ languages including underrepresented dialects. For organisations safeguarding against AI voice fraud, this means continuous, reliable detecti ### MAS highlights three urgent threats for financial institutions in Singapore - Date: 2025-12-10 - URL: https://aurigin.ai/blog/mas-highlights-three-urgent-threats-for-financial-institutions-in-singapore - Summary: Monetary Authority of Singapore (MAS) sounded the alarm on deepfake cyber risks in the financial sector. The report identifies three key threat areas - defeating biometric authentication, impersonation scams, and misinformation. ### Aurigin.ai wins 2nd place at the Swiss Innovation Challenge 2025 - Date: 2025-11-20 - URL: https://aurigin.ai/blog/aurigin-ai-wins-2nd-place-at-the-swiss-innovation-challenge-2025 - Summary: We’re proud to announce that Aurigin.ai has been awarded 2nd place at the 2025 Swiss Innovation Challenge, the largest innovation competition in Switzerland, out of more than 100 projects across three elimination rounds. This recognition means a lot. Over the past eight months, the Swiss Innovation Challenge has brought together some of the most forward-thinking SMEs and startups in the country, supported by the Baselland Chamber of Commerce and FHNW. Being ranked second among such an exception ### Swisscom and Aurigin.ai team up to make KYC faster, safer and fraud proof - Date: 2025-11-12 - URL: https://aurigin.ai/blog/swisscom-and-aurigin-ai-team-up-to-make-kyc-faster-safer-and-fraud-proof - Summary: Swisscom Digital Trust and Aurigin.ai are collaborating to strengthen existing KYC processes with an additional layer of trust and fraud prevention, reinforcing the current framework through voice-based liveness verification and AI deepfake detection.  ### Aurigin.ai joins forces with MCA India’s Deepfakes Analysis Unit to combat misinformation - Date: 2025-10-14 - URL: https://aurigin.ai/blog/aurigin-ai-joins-forces-with-mca-indias-deepfake-analysis-unit-to-combat-misinformation - Summary: We’re pleased to announce a collaboration between Aurigin.ai and the Deepfakes Analysis Unit (DAU) of the Misinformation Combat Alliance (MCA India). A partnership aimed at strengthening the verification of AI-generated and manipulated audio content. Empowering fact-checking with advanced audio forensics The DAU, one of India’s leading fact-checking and deepfake-analysis teams, has integrated Aurigin.ai’s audio deepfake detection engine into its verification workflow. This addition enhances t ### Aurigin.ai keeps 1291 Group’s conversations safe from deepfake impersonation - Date: 2025-09-24 - URL: https://aurigin.ai/blog/aurigin-ai-keeps-1291-groups-conversations-safe-from-deepfake-impersonation - Summary: 1291 Group, global leader in bespoke wealth‑protection solutions has partnered with Aurigin.ai to deploy our real‑time, on‑device deepfake detector across its worldwide workforce.  Deepfake impersonation fraud is exploding  Recent industry research exposes how fast the threat landscape is shifting:  * Deepfake audio fraud losses in financial services are projected to hit USD 5 billion a year by 2026.¹  * Reported deepfake attacks in North America grew 1740 % between 2022 and ### Introducing Apollo, Setting a New Record in Voice Deepfake Detection Accuracy - Date: 2025-08-28 - URL: https://aurigin.ai/blog/apollo-enterprise-ready-deepfake-detection - Summary: “Apollo”, the model we released today, surpasses industry leaders in several dimensions: Its accuracy of 97.7% reduces the error rate of the next-best commercial model by 48.2%. With an ultra-low latency of <50ms it is ideal for real-time applications like securing communication. Starting today, Apollo is available through our API, web app, and enterprise integrations, including on-premises deployments. Closing the Gap in Audio Security Until now, organizations faced a difficult ch ### “My voice is my password”: How deepfakes are triggering a voice authentication fraud crisis - Date: 2025-08-08 - URL: https://aurigin.ai/blog/my-voice-is-my-password-how-deepfakes-are-triggering-a-voice-authentication-fraud-crisis - Summary: In July 2025, OpenAI CEO Sam Altman delivered a stark warning at a Federal Reserve conference: cutting-edge AI voice cloning has gotten so good that criminals could soon trigger a “significant fraud crisis” in banking[1]. Altman noted that some banks allow customers to authenticate themselves over the phone using voice biometrics essentially, “my voice is my password.” He cautioned that attackers will exploit this by using AI to mimic customer voices and authorize large money transfers, bypas ### UBS’ data breach is perfect fuel for deepfake frauds - Date: 2025-06-19 - URL: https://aurigin.ai/blog/ubs-data-breach-is-perfect-fuel-for-deepfake-frauds - Summary: This week, Swiss banking giant UBS confirmed a breach at a third-party supplier that exposed personal details of 130,000 employees, including the direct phone number of CEO Sergio Ermotti. This massive data leak is more than a privacy nightmare; it’s a fraudster’s dream toolkit. Why? Because armed with names, roles, and contact info of key personnel, today’s attackers can go beyond phishing emails; they can call employees impersonating executives using AI-generated voices. Welcome to the era ### Deepfakes are undermining callback verification in private banking - Date: 2025-04-30 - URL: https://aurigin.ai/blog/deepfakes-are-undermining-callback-verification-in-private-banking - Summary: Private banks pride themselves on discretion, trust, and personal relationships. One of the most reliable safeguards against fraud in high-value transactions has been callback verification, a simple practice where a banker or relationship manager calls the client to confirm a large transfer instruction. But in the era of generative AI and deepfakes, even a familiar voice on the phone can no longer be taken at face value. This article explores how callback verification works, why it’s sometime