Arduino Pro Speech Recognition Engine Brings 40+ Languages to Selected Arduino Nano, Portenta Boards
Partnership with Cyberon brings a ready-to-use "zero training" speech recognition engine, with licenses starting at $9 per device.
Arduino has announced another string to its Pro bow, partnering with Cyberon to launch a dedicated speech recognition engine tailored to the Arduino Nano and Portenta families — and landing with the promise of zero training required.
"Voice commands are a contactless, hands-free, natural way to interact with devices, equipment, and machines of all kinds: no wonder they are increasingly popular," says the Arduino team by way of background. "With speech recognition technology advancing faster than you can say 'users expect it,' developers often need to find ways to integrate it in their new projects. This, however, requires them to take the time to train a smart device to listen for wake-up words, understand commands… and potentially start over for every new voice or language."
This, the Arduino team claims, is where its partnership with Cyberon on the Arduino Pro Speech Recognition Engine comes in. Compatible at launch with the Arduino Nano 33 BLE, Arduino Nano RP2040 Connect, and the Arduino Portenta H7 with Portenta Vision Shield or suitable external microphone, the new Speech Recognition Engine comes with the promise of "industrial grade performance."
A big selling point of the tool is the claimed training time of zero. "Because it uses text input to define voice commands," the Arduino team explains, "Speech Recognition Engine translates what you say into instant results, with flexibility over the speaker’s actual voice, tone or accent." That flexibility extends to over 40 languages, comes with automatic filtration of background noises, and offers support for multiple wake-words and word sequences.
As an Arduino Pro product, though, the Speech Recognition Engine isn't free. Arduino offers a free trial which comes with support for a single dataset, a single trigger, up to 20 commands, 50 recognition times, and a 20 second trigger delay. A $9 per-device payment removes the limit on recognition times and the trigger mode delay, while a Pro license removes all limits — at a price yet to be publicly announced.
More information on the Arduino Pro Speech Recognition Engine is available on the Arduino website, along with a link to register for a free trial license.