Porcupine Overview

Porcupine Wake Word
production-ready models, trained in seconds, can run anywhere

Private
All voice data is processed on-device. Intrinsically HIPAA and GDPR compliant.

Zero-Latency
Edge-first architecture eliminates unpredictable network delay.

Reliable
100% of the voice recognition is on-device. No concurrency limit. No downtime.

Free Tier
Train custom models, build, and deploy for free. No credit card required.

Self Service
Train custom models within seconds using Picovoice Console.

Cross Platform
Runs everywhere. Linux, macOS, Windows, Android, iOS, Raspberry Pi, NVIDIA Jetson Nano, BeagleBone, Arm Cortex-M, and modern web browsers are supported.