Saj Link is the application you use every day. Saj Speak, Saj See, and Saj Codec are internal capability modules that surface through Saj Sense Business+, the hosted developer API, and OEM/Embedded engagements — never sold as separate brands.
Post-quantum encrypted communications. Arabic-first. Sovereign deployment.
iOS Build 6 live on TestFlight (Internal + External, review approved), web PWA live at app.sajlink.com, macOS DMG signed and Apple-notarized, Android in development. Backed by Saj Protocol V2 (ML-KEM-768 hybrid key exchange · Triple Ratchet · MLS groups · key transparency).
On-device voice intelligence — transcription, synthesis, wake-word, and more. Multi-engine pipeline.
STT (streaming + batch), TTS, wake-word, voice activity detection, speaker recognition, diariser, intent extraction, and noise suppression. Native platform speech engine bindings on supported macOS and iOS hardware. Voice enrollment and speaker verification available under enterprise NDA. Five Arabic dialects covered at the foundation.
On-device vision — deepfake detection + Arabic OCR. Frames never leave the device.
Deepfake detection, Arabic OCR (RTL-aware, joining script), and scene analysis — all camera-side. Frames never leave the device. Designed for edge silicon in classified and air-gapped environments. Also: object detection, face matching, pose estimation, segmentation.
Proprietary neural audio compression. Arabic-adaptive. Patent-protected.
Multiple quality profiles. Language-adaptive compression tuned for Arabic phonology. Per-frame encrypted audio transport. Watermarking layer for EU AI Act Article 50 provenance. Powers Saj Link voice + provides the OEM/Embedded per-device royalty hook.