
by Sensory
Sensory makes on-device voice, sound and biometrics software for embedded products, including wake words, speech-to-text and speaker verification. Everything runs locally on the device with no cloud dependency.
APIContact for pricing
Step 1: Try the wake word, speech-to-text and phrase spotted command demos on the Sensory site.
Step 2: Request a demo to discuss your product and target hardware.
Step 3: Choose the SDKs you need, such as wake word, speech recognition, biometrics or sound ID.
Step 4: Customize the wake words, commands and recognition models for your product and languages.
Step 5: Integrate the SDK on your target platform and run it on the device.
No. The maker says speech, biometric and sound processing all happen on the device, and no speech or biometric data is uploaded or stored externally.
Yes. On-device processing does not depend on an internet connection, so performance stays consistent with no connectivity.
The site lists Android, iOS, Linux, Windows, RTOS, WASM and leading MCU, DSP and SoC platforms.
Yes. Sensory describes custom, co-branded and user-defined wake words, along with tailored command sets.
Amazon PollyAmazon Polly applies AI to neural text-to-speech service for generating spoken audio in applications, accessibility, and media workflows.
Rev AIRev AI applies AI to speech recognition API for streaming and asynchronous transcription, diarization, and language workflows.