Seeed ReSpeaker now supported by Picovoice: Enabling End-to-End Voice AI on the Edge

Building a unified AI strategy is popular and needy for hardware devices, especially we need edge solutions for protecting privacy. Picovoice provides speech recognition and NLU technologies enabling developers to innovate and differentiate rapidly with private voice AI.

What is Picovoice?

Picovoice is an end-to-end platform for building voice products on your terms. It enables creating voice experiences similar to Alexa and Google Assistant, however, Picovoice runs entirely on-device. Picovoice is

• Private: Everything is processed offline. Intrinsically HIPAA and GDPR compliant.
• Reliable: Runs without needing constant connectivity.
• Zero Latency: Edge-first architecture eliminates unpredictable network delay.
• Accurate: Resilient to noise and reverberation. It outperforms cloud-based alternatives by wide margins *.
• Cross-Platform: Design once, deploy anywhere. Build using familiar languages and frameworks.

Use Picovoice to build edge voice AI solutions with ReSpeaker:

ReSpeaker 4-Mic Array for Raspberry Pi is a quad-microphone expansion board for Raspberry Pi designed for AI and voice applications. Developers can build powerful and flexible voice products that integrate purpose-made and branded voice assistants using Picovoice. These branded voice assistants can co-exist with Amazon Alexa Voice Service and, Google Assistant,, thanks to Picovoice’s open architecture.

This is a simple demo that utilizes the ReSpeaker 4-Mic Array on Raspberry Pi with the Picovoice voice technology to control ReSpeaker LEDs.

This demo is triggered by the pre-defined wake word “Picovoice” and then takes voice commands such as turning ON/OFF LEDs and changing LED colours.

Please follow up the wiki to configure the ReSpeaker with Raspberry Pi before using Picovoice. Through the wiki, we provide the step-by-step tutorial to help you build the edge AI voice device. You can use multiple wake works on the ReSpeaker, which means you can divide the room you want to control into different areas and scenarios.

Demo Source Code

The demo is built with the Picovoice SDK. The demo source code is available on GitHub at https://github.com/Picovoice/picovoice/tree/master/demo/respeaker.

Different wake words

The Picovoice SDK includes free sample wake words licensed under Apache 2.0, including major voice assistants (e.g. “Hey Google”, “Alexa”) and fun ones like “Computer” and “Jarvis”. You can find all free models for Raspberry Pi here. Furthermore, enterprises who are engaged with Picovoice can build custom wake words using Picovoice Console.

We also prepared a multi wake-word example using ReSpeaker 4-Mic Array with Raspberry Pi! Different wake words can set to execute certain tasks.

Custom voice commands

The lighting commands are defined by Picovoice’s Rhino Speech-to-Intent context. You can design and train contexts by typing in the allowed grammar using Picovoice Console. You can test your changes in-browser as you edit with the microphone button. Go to Picovoice Console and sign up for an account. Use the Rhino Speech-to-Intent editor to make contexts, then train them for Raspberry Pi.

We will bring more demos on Picovoice with ReSpeaker. Let us know what demo and what other Seeed’s devices are you expecting in the next stage! We carefully listen to the community and will take action! 

About Author