WisBlock Voice Processing Starter Kit
Voice-enabled WisBlock kit with microphone, speaker, and DSP module
RAK10713
Prefer another courier for your delivery? Contact us for special shipping arrangements.
Check our WisBlock Documentation
What Is The WisBlock Voice Processing Starter Kit?
The WisBlock Voice Processing Starter Kit is an embedded audio development kit and modular development platform for building embedded devices with microphone input, dedicated audio processing, application control, and speaker output.
This WisBlock audio kit combines a WisBlock Core module (RAK3312 or RAK4631), microphone, DSP module, amplifier, speaker, base board, and Unify Enclosure in one system. Developers can use the kit to evaluate local voice interfaces, audio-triggered applications, audible notifications, and voice-enabled IoT concepts without having to source and integrate each audio component separately.
The modular WisBlock architecture also allows compatible sensors and interface modules to be added as application requirements develop.
How the WisBlock Voice Processing Architecture Works
The kit separates audio capture, signal processing, application logic, output, and wireless communication into dedicated layers:

The microphone captures speech or environmental audio. The DSP module processes the incoming signal and passes the resulting command, trigger, or audio data to the WisBlock Core.
The Core module then determines what the device should do. Depending on the application, it can play an audible response, activate another connected component, record an event, or send selected information over a supported wireless connection.
Key Terms Explained
These terms describe related but different functions.
- Voice Processing covers operations that capture, filter, clean, analyze, or modify a spoken audio signal.
- Voice Recognition identifies supported spoken commands or triggers and converts them into events that application software can use.
- Sound Recognition classifies non-speech audio, such as alarms, impacts, machine sounds, or environmental events. Custom sound classification requires a compatible development workflow and trained model.
- Audio Processing is the broader category covering voice, environmental sound, filtering, amplification, playback, and other audio operations.
Supported functions depend on the selected WisBlock Core, firmware, audio libraries, processing model, and example project.
Why Process Voice and Audio Locally?
Local audio processing allows supported voice commands or audio events to be handled on the device rather than continuously sending raw audio to a remote service.
Here are the benefits of processing voice and audio locally:
- Reduced network dependency: The system can process supported commands or audio events without continuously sending raw audio to a remote server.
- Lower response delay: A recognized command can be passed directly to the application without waiting for a cloud round-trip.
- Reduced bandwidth: Only a command result, classification, or event may need to be transmitted instead of the full audio stream.
- Better privacy control: Local processing can reduce the need to transmit raw voice recordings outside the device.
- Offline operation: Supported voice functions can continue when internet access is unavailable, although the exact offline capability depends on the selected model and firmware.
- Predictable system behavior: Local execution reduces variability caused by network congestion, service availability, or cloud latency.
What Can You Build with the WisBlock Voice Processing Starter Kit?
|
Use Case |
Description |
|
Voice-controlled lighting or room controls |
Control lighting or room settings using supported voice commands |
|
Local equipment-control interfaces |
Operate or manage equipment through local voice interaction |
|
Spoken status-request devices |
Request system or sensor status using voice input |
|
Voice-enabled environmental monitoring nodes |
Combine voice interaction with environmental sensing |
|
Audible notification devices on LoRaWAN® |
Receive messages and trigger audible alerts |
|
Local alarm acknowledgement systems |
Acknowledge alarms using voice commands without cloud dependency |
|
Smart kiosk interfaces |
Provide voice-based interaction for kiosks or public terminals |
|
Accessibility control prototypes |
Enable voice-based control for accessibility-focused applications |
|
Audio event monitoring concepts |
Detect and respond to specific audio events or patterns |
|
Educational voice-processing demonstrations |
Demonstrate voice and audio processing concepts in learning environments |
|
Industrial operator interfaces |
Support voice interaction for industrial control or monitoring |
|
Sensor systems with audible responses |
Combine sensor data with audible feedback or alerts |
Which WisBlock Core Module Should You Choose?
The Core module runs application logic and manages the supported wireless communication. Choose the Core based on the application's connectivity, software environment, audio library, and required example projects.
|
Feature |
RAK4631 |
RAK3312 |
|
MCU |
Nordic nRF52840 |
Espressif ESP32-S3 |
|
Architecture |
ARM Cortex-M4F, single-core |
Dual-Core Xtensa LX7 |
|
Clock Speed |
64 MHz |
240 MHz |
|
Flash |
1 MB |
16 MB |
|
RAM |
256 KB SRAM |
8 MB PSRAM |
|
LoRa Transceiver |
Semtech SX1262 |
Semtech SX1262 |
|
LoRaWAN |
Yes |
Yes |
|
BLE |
BLE 5.0 |
BLE 5.0 |
|
WiFi |
No |
Yes (802.11 b/g/n) |
|
Edge AI / ML Acceleration |
Limited |
Yes (vector instructions for DSP & ML) |
|
DSP Module (RAK18080) |
Compatible |
Compatible |
|
WisBlock Audio Library |
Supported |
Supported |
|
Development Framework |
Arduino BSP, RUI3 |
Arduino, ESP-IDF |
|
Mounting Plate |
Built-in LoRa antenna |
Built-in LoRa antenna |
|
Power Consumption |
Ultra-low (optimized for battery) |
Low (higher than nRF52840; supports deep sleep modes) |
|
Best For |
Low-power LoRaWAN audio devices; battery-operated voice-triggered sensors |
Advanced voice/audio processing; AI-at-the-edge (keyword detection, anomaly detection); WiFi-connected audio applications |
Why You Need The WisBlock Voice Processing Starter Kit
Build and evaluate voice-enabled IoT applications faster with this voice processing development kit, designed with hardware that works together. The kit reduces integration effort, helping developers focus on application logic, voice interaction, and testing instead of assembling compatible components.
Why it’s useful:
- Faster Prototyping: Assemble and begin voice application development with fewer integration steps.
- Consistent Development Platform: Uses compatible WisBlock modules for repeatable testing and evaluation.
- Learn Voice Processing: Tutorials and sample code simplify experimentation with audio processing and voice-enabled embedded applications.
Key Features of the WisBlock Voice Processing Starter Kit
- Complete Audio Hardware: Includes a microphone, speaker, and audio amplifier for voice input and output.
- WisBlock Audio Core Options: Available with a supported Audio Core (RAK3312 or RAK4631), allowing developers to choose the software environment that matches their project.
- Dedicated Audio DSP: The DSP module performs dedicated edge audio processing while the WisBlock Core manages application logic and wireless communication.
- WisBlock Modular Platform: Built on the RAK19007 WisBlock Base Board with four sensor slots and one I/O slot for future expansion.
- Audio Interposer Module: Consolidates the audio modules onto a single WisBlock IO slot, simplifying system integration.
- Expandable Base Board: Supports additional compatible WisBlock sensor and interface modules for application-specific development.
- Integrated Speaker: Includes a speaker and amplifier for immediate audio output during development and testing.
- Modular WisBlock Ecosystem: Compatible with the WisBlock platform, making it easy to expand projects with additional sensors, communication modules, and interfaces.
- Ready-to-Assemble Enclosure: Comes with a Unify Enclosure designed for integrated microphone and speaker installation.
- Complete Mounting Hardware: Includes the mounting plate, enclosure, M8 cable, screws, and required accessories for assembling the kit.
- Development Resources Included: Supported by tutorials, documentation, and example code to accelerate prototyping.
Audio Application Development Workflow
After assembling the hardware, developers can configure the software workflow for voice interfaces, audio notifications, sound-event monitoring, or other embedded audio applications.
- Configure Audio Input
Initialize the microphone and define how the application captures speech or environmental sound.
- Configure Audio Processing
Set up the RAK18080 DSP and the required firmware, processing model, or audio parameters for the intended application.
- Build the Application Logic
Program the selected WisBlock Core to interpret the DSP output and determine the required action.
- Define the Response
Configure the application to play an audible notification, activate connected hardware, store an event, or send a wireless message.
- Add Sensors and Connectivity
Expand the application with compatible WisBlock sensors or communication functions for environmental monitoring, equipment status, motion detection, LoRaWAN reporting, Bluetooth interaction, or WiFi connectivity.
- Test and Refine
Evaluate the application using expected voices, sounds, background-noise levels, microphone placement, and operating conditions. Adjust the software, processing model, and response logic as needed.
WisBlock Voice Processing Kit vs Other Voice Development Platforms
|
Feature |
RAKwireless WisBlock Voice Processing Starter Kit (This Product) |
Arduino Nicla Voice |
Seeed ReSpeaker Lite |
Syntiant TinyML Platform |
|
Primary Focus |
Modular voice-enabled IoT development |
TinyML voice and sensor AI |
Voice assistant and embedded audio |
Ultra-low-power voice AI inference |
|
Complete Development Kit |
Yes |
No, Development board only |
Yes |
No, Development board only |
|
Microphone Included |
Yes |
Yes |
Yes |
Yes |
|
Speaker Included |
Yes |
No |
Yes |
No |
|
Dedicated DSP for Audio |
Yes |
AI processor instead of DSP |
Yes |
Dedicated neural processor |
|
Local Audio Processing |
Yes |
Yes |
Yes |
Yes |
|
Modular Sensor Expansion |
Extensive WisBlock ecosystem |
Limited external expansion |
Limited external expansion |
Limited external expansion |
|
Wireless IoT Connectivity |
LoRaWAN, LoRa, BLE (depending on Core) |
BLE |
WiFi |
Depends on the host platform |
|
Ready for Remote IoT Applications |
Yes |
Limited |
Limited |
Limited |
|
Enclosure Included |
Yes |
No |
Yes |
No |
|
Audio Input + Output Platform |
Yes |
Primarily audio input |
Yes |
Primarily audio input |
|
Expandable Hardware Platform |
Yes |
Limited |
Moderate |
Limited |
|
Edge AI / TinyML Support |
Supported through compatible workflows |
Native |
Supported |
Native |
|
Open Development Environment |
Arduino & ESP-IDF |
Arduino |
ESP-IDF / ESPHome |
SDK |
|
Best For |
Voice-enabled IoT devices, remote monitoring, and industrial prototyping |
TinyML voice recognition |
Smart speakers, Home Assistant, voice assistants |
Ultra-low-power voice AI products |
Supported Software and Development Workflows
Workflow 1: DSP-Based Voice Processing
In this workflow, the microphone captures the user’s voice and the DSP performs the supported processing. The resulting command or event is passed to the WisBlock Core, which runs the programmed action.
Use this workflow for defined voice-control interfaces based on a compatible DSP firmware, voice model, and Core module.
Workflow 2: Sound Recognition Development
A compatible machine-learning workflow can be used to collect audio samples, train a model, and classify selected speech or non-speech sounds.
This approach is appropriate for application-specific sounds such as alarms, machine conditions, impacts, or environmental events. Core support, memory requirements, and model compatibility must be verified before development.
Workflow 3: Audio Input and Output Development
Developers can capture microphone input, process or route audio, and produce speaker output without implementing voice recognition. This workflow supports audio experiments, notifications, playback, and general embedded audio development.
Why Choose the WisBlock Voice Processing Starter Kit?
The WisBlock Voice Processing Starter Kit is a modular development platform for building embedded devices that capture audio, process supported voice commands or sound events, run application logic, and provide an audible or wireless response.
Unlike standalone voice development boards that focus only on audio processing, it combines microphone input, DSP processing, speaker output, modular sensor expansion, and IoT connectivity into a single platform.
Build voice-enabled IoT devices, embedded audio applications, local voice processing systems, and audio-enabled remote monitoring solutions using compatible WisBlock sensors and communication modules. Its modular architecture simplifies prototyping while providing a scalable path toward real-world IoT deployments.
What’s in the Box
-
RAK4631 Core module Variant:
- 1 pc RAK19007 WisBlock Base Board
- 1 pc RAK4631 WisBlock Core Module
- 1 pc RAK18003 Audio Interposer Module
- 1 pc RAK18030 Microphone Module
- 1 pc RAK18080 DSP module
- 1 pc RAK18061 Amplifier Module
- 1 pc Unify Enclosure prepared for the microphone and speaker
- 1 pc Integrated Antenna and Screw Boosses Mounting Plate for Unify Enclosure
- 1 pc M8 Cable
- 2 pc Spacer
- 1 pc USB Cable
- 1 pc FPC Cable
- 1 set screw
-
RAK3312 Core module Variant:
- 1 pc RAK19007 WisBlock Base Board
- 1 pc RAK3312 WisBlock Core Module
- 1 pc RAK18003 Audio Interposer Module
- 1 pc RAK18030 Microphone Module
- 1 pc RAK18080 DSP module
- 1 pc RAK18061 Amplifier Module
- 1 pc Unify Enclosure prepared for the microphone and speaker
- 1 pc Screw Boosses Mounting Plate for Unify Enclosure
- 1 pc M8 Cable
- 2 pc Spacer
- 1 pc USB Cable
- 1 pc FPC Cable
- 1 set screw
- 1 pc PCB Antenna for LoRa (MHF4 connector)
- 1 pc 2.4GHz Wi Fi/BLE PCB Antenna (MHF4 Connector)
How to Get Started with the WisBlock Voice Processing Starter Kit
- Quick Start Guide for WisBlock
- Source Code Repository for WisBlock
- Download 40-Pin Male Connector 3D File
- Build Your Own Voice Recognition Device Using Wisblock Audio
- How to Make Your Own WisBlock IO Board
- Datasheet for RAK4631 WisBlock LoRaWAN Module
- Datasheet for RAK3312 WisBlock LoRaWAN Module
- Datasheet for RAK19007 WisBlock Base Board 2nd Gen
- Datasheet for RAK18003 WisBlock Audio Interposer Module
- Datasheet for RAK18061 WisBlock Audio Mono Amplifier Module
- Datasheet for RAK18080 WisBlock Audio DSP Board Module
Frequently Asked Questions
Questions and Answers
Didn't find what you're looking for?
Submit your questions and our team will get back to you soon.
Product Reviews
- Quick Start Guide for WisBlock
- Source Code Repository for WisBlock
- Download 40-Pin Male Connector 3D File
- Build Your Own Voice Recognition Device Using Wisblock Audio
- How to Make Your Own WisBlock IO Board
- Datasheet for RAK4631 WisBlock LoRaWAN Module
- Datasheet for RAK3312 WisBlock LoRaWAN Module
- Datasheet for RAK19007 WisBlock Base Board 2nd Gen
- Datasheet for RAK18003 WisBlock Audio Interposer Module
- Datasheet for RAK18061 WisBlock Audio Mono Amplifier Module
- Datasheet for RAK18080 WisBlock Audio DSP Board Module
Flexible Payments with PayPal Pay Later