Sample Applications

The following applications demonstrate the use of the TensorFlow Lite for Microcontrollers framework with the Gecko SDK.

Voice Control Light

This application demonstrates a neural network with TensorFlow Lite for Microcontrollers to detect the spoken words "on" and "off" from audio data recorded on the microphone in a Micrium OS kernel task.

The detected keywords are used to control an LED on the board. The audio data is sampled continuously and preprocessed using the Audio Feature Generator component. Inference is run every 200 ms on the past ~1 s of audio data.

This sample application uses the Flatbuffer Converter Tool to add the .tflite file to the application binary.

Z3SwitchWithVoice

This application combines voice detection with Zigbee 3.0 to create a voice-controlled switch node that can be used to toggle a light node. The application uses the same model as Voice Control Light to detect the spoken keywords "on" and "off". Upon detection, the switch node sends On/Off commands over the Zigbee network.

This sample application uses the Flatbuffer Converter Tool to add the .tflite file to the application binary.

TensorFlow Lite Micro - Hello World

This application demonstrates a model trained to replicate a sine function and use the inference results to fade an LED. The application is originally written by TensorFlow, but has been ported to the Gecko SDK.

The model is approximately 2.5 KB. The entire application takes around 157 KB flash and 15 KB RAM. This application uses large amounts of flash memory because it does not manually specify which operations are used in the model and, as a result, compiles all kernel implementations.

The application illustrates a minimal inference application and serves as a good starting point for understanding the TensorFlow Lite for Microcontrollers model interpretation flow.

This sample application uses a fixed model contained in hello_world_model_data.cc.

TensorFlow Lite Micro - Micro Speech

This application demonstrates a 20 KB model trained to detect simple words from speech data recorded from a microphone. The application is originally written by TensorFlow, but has been ported to the Gecko SDK.

This application uses around 100 KB flash and 37 KB of RAM. Around 10 KB of the RAM usage is related to FFT frontend and to store audio data. With a clock speed of 38.4 MHz and using the optimized kernel implementations, the inference time on ~1 s of audio data is approximately 111 ms.

This application illustrates the process of generating features from audio data and doing detections in real time. It also demonstrates how to manually specify which operations are used in the network, which saves a significant amount of flash.

This sample application uses a fixed model contained in micro_speech_model_data.cc.

TensorFlow Lite Micro - Magic Wand

This application demonstrates a 10 KB model trained to recognize various hand gestures using an accelerometer to detect the motion. The detected gestures are printed to the serial port. The application is originally written by TensorFlow, but has been ported to the Gecko SDK.

This application uses around 104 KB flash and 25 KB of RAM. This application demonstrates how to use accelerometer data as inference input and also shows how to manually specify which operations are used in the network, which saves a significant amount of flash.

This sample application uses the Flatbuffer Converter Tool to add the .tflite file to the application binary.

TensorFlow Model Profiler

This application is designed to profile a TensorFlow Lite Micro model on Silicon Labs hardware. The model used by the application is provided by a TensorFlow Lite flatbuffer file called model.tflite in the config/tflite subdirectory. The profiler will measure the number of CPU clock cycles and elapsed time in each layer of the model when performing an inference. It will also produce a summary when inference is done. The input layer of the model is filled with all zeroes before performing a single inference. Profiling results are transmitted over VCOM.

To run the application with a different .tflite model, you can replace the file called model.tflite with a new TensorFlow Lite Micro flatbuffer file. This new file must also be called "model.tflite" and be placed inside the config/tflite subdirectory to be picked up by the sample application. After the model has been replaced, regenerate the project.

To load and perform inference on a TensorFlow Lite Micro model, allocate a number of bytes to a "tensor arena" to hold state needed by the TensorFlow Lite Micro. The size of this tensor arena depends on the size of the model and the number of operators. The TensorFlow Model Profiler application can be used to measure the amount of RAM needed by the tensor arena to load the specific TensorFlow Lite Micro model. This is measured by dynamically allocating RAM for the tensor arena and reporting the number of bytes needed on VCOM. The number of bytes needed for the tensor arena can later be used to statically allocate memory when the model is used in a different application.

This sample application uses the Flatbuffer Converter Tool to add the .tflite file to the application binary.