RoadTest: Enroll Now to Review the Arduino Nicla Voice
Author: rahulkhanna
Creation date:
Evaluation Type: Connectors & Cable
Did you receive all parts the manufacturer stated would be included in the package?: True
What other parts do you consider comparable to this product?: Generic Voice-detection module
What were the biggest problems encountered?: Limited resources for the Hardware.
Detailed Review:
The Arduino Nicla Voice is a tiny powerhouse, packing speech and motion recognition into a compact 22.86mm x 22.86mm board. Let's dive into what this board offers and how it performs.
The Nicla Voice arrives in a small, professional package. Inside, the board is well-protected and comes with a brief getting-started guide. The board itself is impressive; small yet well-built, with clean soldering and clear labelling.
The heart of the Nicla Voice is the Syntiant NDP120 Neural Decision Processor. This chip allows for efficient on-device machine-learning tasks. The board also boasts a microphone, a 6-axis motion sensor, a magnetometer, and Bluetooth Low-Energy (BLE) connectivity. The Arduino IDE provides libraries and examples to get you started quickly.
Compared to standard Arduino modules, the Nicla Voice offers a significant leap in functionality. Integrating voice and motion recognition opens doors for entirely new project possibilities. While some add-on modules offer similar features, the Nicla Voice's small size and on-board processing make it a more streamlined solution.
| Feature | Arduino Nicla Voice | Seed Studio Grove Voice Recognition Module |
Gravity: Voice Recognition Module |
VOICE RECOGNITION MODULE V3.1 |
| Microcontroller (MCU) | nRF52832 (Cortex-M4) | Nuvoton ISD9160CFI (Cortex-M0+) | n/a | n/a |
| AI Processor | Syntiant NDP120 | None | None | None |
| Onboard Sensors | Microphone, 6-axis IMU, Magnetometer | Microphone | Microphone | External Microphone |
| Connectivity | Bluetooth Low Energy (BLE) | I2C | I2S | UART |
| Power Consumption | Ultra-low power, battery operable | Moderate power | 3.3 - 5V, ≤370 mA | 5V, <40mA |
| Speech Recognition | Always-on, custom keyword spotting | Limited commands |
Requires cloud processing (limited commands) |
Limited commands |
| Model Training | Supports on-device training with Edge Impulse | Pre-trained model can recognize 22 pieces of commands |
121 built-in fixed command words and supports the addition of 17 custom command words (Cloud-based training) |
On-device training upto maximum 80 voice commands |
| Size | 22.86 x 22.86 mm | 45 x 20 mm | 49 x 32 mm | 31 x 50 mm |
| Applications | Wearables, predictive maintenance, gesture recognition, contactless interfaces | Simple voice commands (e.g., turn on/off) |
Simple voice commands (e.g., turn on/off) |
Simple voice commands (e.g., turn on/off) |
| Development Platform | Arduino IDE, Edge Impulse | - | Arduino IDE | - |
Our first project involved keyword detection. We will set up the Nicla Voice on the Arduino IDE, and update the syntiant firmware to run the program. The Nicla Voice effectively recognized the target keyword even in moderately noisy environments.
We built a simple project using Edge Impulse to classify cry sounds and noises.
Edge Impulse is a fantastic platform for training machine learning models on the Nicla Voice. The user interface is intuitive, and pre-built libraries for audio classification and keyword spotting make it easy to start. Uploading the trained model to the board is a smooth process.
The process was straightforward, and the Nicla Voice performed admirably, accurately identifying the sounds with minimal training data.
Review of Documentation, and User Manual: The official documentation for the Nicla Voice is comprehensive and well-written. It covers everything from setting up the board to using its various functionalities. The basic project tutorials are clear and concise, providing a good starting point for beginners.
The Arduino Nicla Voice is a game-changer for edge computing projects. Its small size, powerful processing, and ease of use with Edge Impulse make it a valuable tool for makers and developers alike. While the price point might be slightly higher than standard Arduino modules, the added functionalities justify the cost. If you're looking to add voice and motion control to your industrial projects, the Nicla Voice is an excellent choice.
