Aurigin AI is a deep-tech security company focused on anti-spoofing and liveness detection for voice biometric systems. The company develops AI-based audio deepfake detection technology designed to identify synthetic, cloned, and AI-generated voices in real time. Products and Integration: - REST API: A single HTTP endpoint returning liveness verdicts in under 50ms - Native SDKs: Libraries for Python, TypeScript, Swift, Node.js, and Java - On-Premise deployment: Docker and Kubernetes options for full data sovereignty Key technical capabilities include: - Detection of text-to-speech (TTS) synthetic voices, voice conversion (VC), replay attacks, and personalized voice clones - Support for 80+ languages with a language-agnostic neural architecture - Compatibility with audio ranging from 8 kHz narrowband to 48 kHz WebRTC - Calibrated confidence scores with segment-level localization rather than binary verdicts - Continuous model retraining against emerging synthesis engines such as ElevenLabs, XTTS, RVC, and Bark The company targets voice biometric providers, contact center platforms, and authentication solution providers. Their latest model, Apollo V4, is trained on over 1 million hours of audio data and claims 98%+ accuracy across benchmark datasets. Compliance and privacy commitments include GDPR, SOC 2, and ISO 27001. Audio processed via the cloud API is deleted immediately after analysis and is not used for training. The company is identified as Swiss-made software based on a footer badge on their website.