Question/Issue:
I have a Sound Event Classification model created on EdgeImpulse able to detect the sound of a Toilet Flush from the rest (Noise). I have deployed the model as an Arduino .zip library and imported it in the Arduino IDE. With some compatibility issues, I finally got it uploaded on my Arduino Nano ESP 32. However, the problem now is that my model should do a real-time continuous Classification, but when I upload the sketch, on the serial monitor I see nothing. The Arduino already has an INMP-441 i2s microphone connected to it.
The code of the sketch is as follows:
/* Edge Impulse Arduino examples
* Copyright (c) 2022 EdgeImpulse Inc.
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to deal
* in the Software without restriction, including without limitation the rights
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
// These sketches are tested with 2.0.4 ESP32 Arduino Core
// https://github.com/espressif/arduino-esp32/releases/tag/2.0.4
// If your target is limited in memory remove this macro to save 10K RAM
#define EIDSP_QUANTIZE_FILTERBANK 0
/*
** NOTE: If you run into TFLite arena allocation issue.
**
** This may be due to may dynamic memory fragmentation.
** Try defining "-DEI_CLASSIFIER_ALLOCATION_STATIC" in boards.local.txt (create
** if it doesn't exist) and copy this file to
** `<ARDUINO_CORE_INSTALL_PATH>/arduino/hardware/<mbed_core>/<core_version>/`.
**
** See
** (https://support.arduino.cc/hc/en-us/articles/360012076960-Where-are-the-installed-cores-located-)
** to find where Arduino installs cores on your machine.
**
** If the problem persists then there's not enough memory for this model and application.
*/
/* Includes ---------------------------------------------------------------- */
#include <WLB_TwoClasses_inferencing.h>
#include "freertos/FreeRTOS.h"
#include "freertos/task.h"
#include "driver/i2s.h"
/** Audio buffers, pointers and selectors */
typedef struct {
signed short *buffers[2];
unsigned char buf_select;
unsigned char buf_ready;
unsigned int buf_count;
unsigned int n_samples;
} inference_t;
static inference_t inference;
static const uint32_t sample_buffer_size = 2048;
static signed short sampleBuffer[sample_buffer_size];
static bool debug_nn = false; // Set this to true to see e.g. features generated from the raw signal
static int print_results = -(EI_CLASSIFIER_SLICES_PER_MODEL_WINDOW);
static bool record_status = true;
/**
* @brief Arduino setup function
*/
void setup()
{
// put your setup code here, to run once:
Serial.begin(115200);
// comment out the below line to cancel the wait for USB connection (needed for native USB)
while (!Serial);
Serial.println("Edge Impulse Inferencing Demo");
// summary of inferencing settings (from model_metadata.h)
ei_printf("Inferencing settings:\n");
ei_printf("\tInterval: ");
ei_printf_float((float)EI_CLASSIFIER_INTERVAL_MS);
ei_printf(" ms.\n");
ei_printf("\tFrame size: %d\n", EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE);
ei_printf("\tSample length: %d ms.\n", EI_CLASSIFIER_RAW_SAMPLE_COUNT / 16);
ei_printf("\tNo. of classes: %d\n", sizeof(ei_classifier_inferencing_categories) / sizeof(ei_classifier_inferencing_categories[0]));
run_classifier_init();
ei_printf("\nStarting continious inference in 2 seconds...\n");
ei_sleep(2000);
if (microphone_inference_start(EI_CLASSIFIER_SLICE_SIZE) == false) {
ei_printf("ERR: Could not allocate audio buffer (size %d), this could be due to the window length of your model\r\n", EI_CLASSIFIER_RAW_SAMPLE_COUNT);
return;
}
ei_printf("Recording...\n");
}
/**
* @brief Arduino main function. Runs the inferencing loop.
*/
void loop()
{
bool m = microphone_inference_record();
if (!m) {
ei_printf("ERR: Failed to record audio...\n");
return;
}
signal_t signal;
signal.total_length = EI_CLASSIFIER_SLICE_SIZE;
signal.get_data = µphone_audio_signal_get_data;
ei_impulse_result_t result = {0};
EI_IMPULSE_ERROR r = run_classifier_continuous(&signal, &result, debug_nn);
if (r != EI_IMPULSE_OK) {
ei_printf("ERR: Failed to run classifier (%d)\n", r);
return;
}
if (++print_results >= (EI_CLASSIFIER_SLICES_PER_MODEL_WINDOW)) {
// print the predictions
ei_printf("Predictions ");
ei_printf("(DSP: %d ms., Classification: %d ms., Anomaly: %d ms.)",
result.timing.dsp, result.timing.classification, result.timing.anomaly);
ei_printf(": \n");
for (size_t ix = 0; ix < EI_CLASSIFIER_LABEL_COUNT; ix++) {
ei_printf(" %s: ", result.classification[ix].label);
ei_printf_float(result.classification[ix].value);
ei_printf("\n");
}
#if EI_CLASSIFIER_HAS_ANOMALY == 1
ei_printf(" anomaly score: ");
ei_printf_float(result.anomaly);
ei_printf("\n");
#endif
print_results = 0;
}
}
static void audio_inference_callback(uint32_t n_bytes)
{
for(int i = 0; i < n_bytes>>1; i++) {
inference.buffers[inference.buf_select][inference.buf_count++] = sampleBuffer[i];
if(inference.buf_count >= inference.n_samples) {
inference.buf_select ^= 1;
inference.buf_count = 0;
inference.buf_ready = 1;
}
}
}
static void capture_samples(void* arg) {
const int32_t i2s_bytes_to_read = (uint32_t)arg;
size_t bytes_read = i2s_bytes_to_read;
while (record_status) {
/* read data at once from i2s */
i2s_read((i2s_port_t)1, (void*)sampleBuffer, i2s_bytes_to_read, &bytes_read, 100);
if (bytes_read <= 0) {
ei_printf("Error in I2S read : %d", bytes_read);
}
else {
if (bytes_read < i2s_bytes_to_read) {
ei_printf("Partial I2S read");
}
// scale the data (otherwise the sound is too quiet)
for (int x = 0; x < i2s_bytes_to_read/2; x++) {
sampleBuffer[x] = (int16_t)(sampleBuffer[x]) * 8;
}
if (record_status) {
audio_inference_callback(i2s_bytes_to_read);
}
else {
break;
}
}
}
vTaskDelete(NULL);
}
/**
* @brief Init inferencing struct and setup/start PDM
*
* @param[in] n_samples The n samples
*
* @return { description_of_the_return_value }
*/
static bool microphone_inference_start(uint32_t n_samples)
{
inference.buffers[0] = (signed short *)malloc(n_samples * sizeof(signed short));
if (inference.buffers[0] == NULL) {
return false;
}
inference.buffers[1] = (signed short *)malloc(n_samples * sizeof(signed short));
if (inference.buffers[1] == NULL) {
ei_free(inference.buffers[0]);
return false;
}
inference.buf_select = 0;
inference.buf_count = 0;
inference.n_samples = n_samples;
inference.buf_ready = 0;
if (i2s_init(EI_CLASSIFIER_FREQUENCY)) {
ei_printf("Failed to start I2S!");
}
ei_sleep(100);
record_status = true;
xTaskCreate(capture_samples, "CaptureSamples", 1024 * 32, (void*)sample_buffer_size, 10, NULL);
return true;
}
/**
* @brief Wait on new data
*
* @return True when finished
*/
static bool microphone_inference_record(void)
{
bool ret = true;
if (inference.buf_ready == 1) {
ei_printf(
"Error sample buffer overrun. Decrease the number of slices per model window "
"(EI_CLASSIFIER_SLICES_PER_MODEL_WINDOW)\n");
ret = false;
}
while (inference.buf_ready == 0) {
delay(1);
}
inference.buf_ready = 0;
return true;
}
/**
* Get raw audio signal data
*/
static int microphone_audio_signal_get_data(size_t offset, size_t length, float *out_ptr)
{
numpy::int16_to_float(&inference.buffers[inference.buf_select ^ 1][offset], out_ptr, length);
return 0;
}
/**
* @brief Stop PDM and release buffers
*/
static void microphone_inference_end(void)
{
i2s_deinit();
ei_free(inference.buffers[0]);
ei_free(inference.buffers[1]);
}
static int i2s_init(uint32_t sampling_rate) {
// Start listening for audio: MONO @ 8/16KHz
i2s_config_t i2s_config = {
.mode = (i2s_mode_t)(I2S_MODE_MASTER | I2S_MODE_RX | I2S_MODE_TX),
.sample_rate = sampling_rate,
.bits_per_sample = (i2s_bits_per_sample_t)16,
.channel_format = I2S_CHANNEL_FMT_ONLY_RIGHT,
.communication_format = I2S_COMM_FORMAT_I2S,
.intr_alloc_flags = 0,
.dma_buf_count = 8,
.dma_buf_len = 512,
.use_apll = false,
.tx_desc_auto_clear = false,
.fixed_mclk = -1,
};
i2s_pin_config_t pin_config = {
.bck_io_num = 18, // IIS_SCLK
.ws_io_num = 10, // IIS_LCLK
.data_out_num = -1, // IIS_DSIN
.data_in_num = 17, // IIS_DOUT
};
esp_err_t ret = 0;
ret = i2s_driver_install((i2s_port_t)1, &i2s_config, 0, NULL);
if (ret != ESP_OK) {
ei_printf("Error in i2s_driver_install");
}
ret = i2s_set_pin((i2s_port_t)1, &pin_config);
if (ret != ESP_OK) {
ei_printf("Error in i2s_set_pin");
}
ret = i2s_zero_dma_buffer((i2s_port_t)1);
if (ret != ESP_OK) {
ei_printf("Error in initializing dma buffer with 0");
}
return int(ret);
}
static int i2s_deinit(void) {
i2s_driver_uninstall((i2s_port_t)1); //stop & destroy i2s driver
return 0;
}
#if !defined(EI_CLASSIFIER_SENSOR) || EI_CLASSIFIER_SENSOR != EI_CLASSIFIER_SENSOR_MICROPHONE
#error "Invalid model for current sensor."
#endif
Project ID:
The ID of my EdgeImpulse project is 1001180
Steps Taken:
- Upload the sketch
- Verify nothing happens
- [Step 3]
Expected Outcome:
I should (or would like to) see on the serial monitor the live result of the classification of the audio feeded from the microphone to the board in real-time.
Actual Outcome:
the serial monitor is empty
Reproducibility:
- [ ] Always
Environment:
- **Platform:**Arduino Nano ESP32
- Build Environment Details: Arduino IDE 2.3.8
- **OS Version: Ubuntu 20.04
- Edge Impulse Version (Firmware): [e.g., 1.2.3]
- To find out Edge Impulse Version:
- if you have pre-compiled firmware: run edge-impulse-run-impulse --raw and type AT+INFO. Look for Edge Impulse version in the output.
- if you have a library deployment: inside the unarchived deployment, open model-parameters/model_metadata.h and look for EI_STUDIO_VERSION_MAJOR, EI_STUDIO_VERSION_MINOR, EI_STUDIO_VERSION_PATCH
- Edge Impulse CLI Version: [e.g., 1.5.0]
- Project Version: [e.g., 1.0.0]
-
Custom Blocks / Impulse Configuration: the impulse has a MFE learning block and a NN classifier
Logs/Attachments:
→Impulse configuration
Additional Information:
[Any other information that might be relevant]