Speech to Intent

The world's most efficient NLU engine that can run even on microcontrollers

Accurate

Accurate

Resilient to noise, reverberation, and works across a variety of accents. Outperforms even cloud-based solutions such as Google's Dialogflow width high margins [ 1 ]

Realtime

Realtime

All processing is done on-device in realtime with minimum-possible latency.

60 milliseconds
Picovoice's latency on an ARM Cortex-M4
2100 milliseconds
Dialogflow's latency
Lightweight

Lightweight

Built upon proprietary deep learning technology that runs on embedded hardware efficiently. Can run even on small microcontrollers such as ARM Cortex-M4 in realtime.

80 MCPS
CPU usage on ARM Cortex-M4
90 KB
RAM
300 KB
ROM
Demo (ARM Cortex-M4)      Demo (ARM Cortex-M7)     
Cross Platform

The only voice AI platform that scales from microcontrollers to web browsers

ARM
NXP Semiconductors
STMicroelectronics
Raspberry Pi
Texas Instruments
Android
iOS
FireFox
Chrome
Linux
macOS
Windows
Request Access      Use Picovoice Console      Get Started on GitHub     

Live Demo - Speech to Intent Engine

This demo works only on Firefox and Chrome browsers. Additionally, you need a working microphone.

This demo allows you to issue voice commands to a smart coffee maker. Note that you need a working microphone. You can ask the coffee maker to make you a drink using voice commands such as

  • Hey Barista, can I have a latte?
  • Hey Barista, make me a single-shot espresso.
  • Hey Barista, I want a triple-shot americano with milk.
  • Hey Barista, may I have a large cappuccino with cream?

Small   Medium   Large  

Single Shot   Double Shot   Triple Shot  

Americano   Cappuccino   Espresso   Latte   Mocha  

Milk   Cream  

coffee maker

Contact Enterprise Team