Sensory Smart Wake Word

0 reviews

Sensory Smart Wake Word is part of Sensory Inc.’s on‑device voice AI portfolio, developed by a long‑standing leader in embedded wake word and speech technologies used across billions of devices. Building on years of experience powering “Hey Siri,” “OK Google,” and other branded triggers through its TrulyHandsfree engine, Sensory now extends that expertise into next‑generation, context‑aware wake words for modern voice assistants and LLM‑driven experiences. Sensory Smart Wake Word is a new generation of wake word technology that uses on‑device AI to understand not just what phrase was spoken, but whether the device should actually respond. Unlike rigid, single‑phrase triggers, Smart Wake Words can adapt sensitivity based on context, noise, and user behavior, support more natural follow‑up speech without forcing the user to repeat the wake phrase, and optionally tie activation to a specific user’s voice for added personalization or security. The models run at ultra‑low power on edge hardware and can be combined with Sensory’s STT, NLU, biometrics, and SoundID to form the front door for LLM voice agents in automotive, consumer, and enterprise products. For product teams, Smart Wake Word reduces false activations and missed triggers while making voice interactions feel more natural—especially in noisy, shared, or far‑field environments. It lets OEMs move beyond basic “always the same phrase” behavior to more intelligent, context‑aware activation that respects privacy by running entirely on‑device and only escalating to cloud or LLM checks when appropriate. This helps brands deliver more intuitive, trustworthy voice UX in cars, homes, and devices, while maintaining full control over the trigger phrase, wake behavior, and overall assistant experience.

Sensory Personalized Wake Word

0 reviews

Sensory Personalized Wake Word is an extension of Sensory Inc.’s on‑device wake word portfolio, designed to give end users their own custom activation phrases while leveraging the same embedded AI technology trusted by leading OEMs. It brings Sensory’s decades of wake word and voice biometric innovation into a user‑centric experience that works across languages, accents, and environments. Personalized Wake Word lets individuals define and enroll their own wake words in under a minute, typically with just a handful of spoken repetitions, and then use that phrase to wake the device going forward. The models run fully on‑device, support virtually any phrase, including invented or non‑dictionary words, and can optionally be tied to the user’s voiceprint for added discrimination on shared devices. Enrollment and recognition are optimized for low power, far‑field, and noisy conditions, and can be configured and prototyped through Sensory VoiceHub for a range of hardware platforms. For end users, Personalized Wake Word makes devices feel more personal and accessible, replacing generic triggers with phrases that fit their preferences, language, or brand relationship. For OEMs and solution providers, it offers a way to differentiate products, support multi‑user environments, and introduce a baseline of security by filtering out non‑enrolled speakers, without storing audio in the cloud or relying on network connectivity. This is especially valuable for shared home devices, vehicles, and enterprise hardware where personalization, privacy, and offline reliability all matter.

Sensory Secure Wake Word

0 reviews

Sensory Secure Wake Word sits at the intersection of Sensory’s wake word and voice biometric technologies, offering on‑device voice authentication from a company recognized for embedded AI in security‑sensitive applications. It builds on Sensory's Speaker Verification and long history of wake word deployments to provide a wake phrase that doubles as a biometric security factor. Secure Wake Word combines wake word detection with built‑in, text‑dependent speaker verification so devices only wake when the correct enrolled user says the trigger phrase. During a short enrollment process, the user repeats their chosen phrase a few times while Sensory creates a compact voiceprint tied to that phrase; from then on, the device continuously listens at ultra‑low power and wakes only when both the phrase and voice match. Multiple users can enroll separately on shared devices, and OEMs can choose between branded fixed phrases, user‑defined phrases, or a combination, with security implemented either directly in the wake word layer or as a second‑stage check at the OS or application level. Secure Wake Word lets product teams build hands‑free experiences that are not just convenient but also resistant to casual misuse, replay, or accidental activation by non‑owners. Because all processing runs locally, organizations can strengthen access control and personalization while keeping biometric data on the device and avoiding recurring cloud authentication costs. This makes Secure Wake Word a strong fit for devices and applications where both UX and security are critical, such as smart home controllers, automotive systems, enterprise endpoints, and financial or health‑related apps.

Sensory Speech-to-Text

0 reviews

Real-time transcription that runs accurately on modern operating systems and chipsets, with no cloud dependency or metered fees and no compromise on privacy - speech-to-text that you can trust anywhere.

Sensory Phrase Spotted Commands

0 reviews

Recognize multiple voice commands at once, respond in real time, and keep everything running fully on-device and in low power with minimal memory.