Hi. How can i use micropython firmware alongside a Max9814 , with this module? Any Solutions?

I have written the code below but cant hear clear voice in audacity...

from machine import Pin, ADC
import ustruct , time

analog_value = machine.ADC(26)
conversion_factor =3.3/(65536)

samples = []

while True:
    reading = analog_value.read_u16()*conversion_factor   
samples.append(int(reading)) #print("ADC: ",reading)
time.sleep(0.002)



with open('Voice.bin', 'wb') as output:
     for sample in samples:
        output.write(struct.pack('<h', sample))

Thanks to this project, make me able to make my own microphone with excellent sound quality without having to write code.

At first time, I tested many microphones, SPK0838HT4H has ~ -45dB background noise, MSM26D4040H11 has ~ -35dB, MSM261D4030H1C has ~ 33dB. I checked the DCDC power supply. and ask the resellers why those microphones have so much noise but no useful result

when I decide to port RNNoise to PI2040, I want to add a simple -25dB suppressor first. then I find this...

https://github.com/ArmDeveloperEcosystem/microphone-library-for-pico/blob/a837f633a6ad2bac268349df35d57e46e551f416/src/pdm_microphone.c#L96-L103

after add pdm_microphone_set_filter_gain(1); to the main function in examples/usb_microphone/main.c all microphones works perfect with < -90dB noise.

So please consider changing the default gain to 1.

* a picture of my microphone

Hi, I was trying to record some audio using the same microphone setup as you instructed. It was a loud audio and for the same reason clipping was observed when the recording was analyzed. I tried to change the "Filter Gain" and "Volume" parameters to reduce the gain but didn't work. It would be great to have some lead on reducing the input gain. Thanks!!

Hi, thank you for the sample code. I tried it on Ubuntu 18.04 and Windows. While Ubuntu 18.04 finds MicNode and shows it in Audacity, the device fails to start in Windows 10.

• MicNode does appear under "Sound, video and game devices" in Device Manager, but with a yellow triangle icon (apparently indicating that the device failed to start)
• right-clicking on MicNode in Device Manager shows device status as "This device cannot start. (Code 10) The specified range could not be found in the range list."

Any advice would be much appreciated. Thank you

I made the following changes to the main.c file in the hello_analog_microphone sample

/*
 * Copyright (c) 2021 Arm Limited and Contributors. All rights reserved.
 *
 * SPDX-License-Identifier: Apache-2.0
 * 
 * 
 * This examples captures data from an analog microphone using a sample
 * rate of 8 kHz and prints the sample values over the USB serial
 * connection.
 */

#include <stdio.h>

#include "pico/stdlib.h"
#include "pico/analog_microphone.h"
#include "tusb.h"

#define AUDIO_SAMPLE_RATE 8000
#define RECORD_LENGTH_S   10

// configuration
const struct analog_microphone_config config = {
    // GPIO to use for input, must be ADC compatible (GPIO 26 - 28)
    .gpio = 26,

    // bias voltage of microphone in volts
    .bias_voltage = 0.0,

    // sample rate in Hz
    // .sample_rate = 8000,
    .sample_rate = AUDIO_SAMPLE_RATE,

    // number of samples to buffer
    .sample_buffer_size = 256,
};

// variables
int16_t sample_buffer[256];
volatile int samples_read = 0;

int16_t full_record[AUDIO_SAMPLE_RATE * RECORD_LENGTH_S]; 
int full_record_INDEX = 0;

void on_analog_samples_ready()
{
    // callback from library when all the samples in the library
    // internal sample buffer are ready for reading 
    samples_read = analog_microphone_read(sample_buffer, 256);
}

int main( void )
{
    bool break_while = false;

    // initialize stdio and wait for USB CDC connect
    stdio_init_all();
    while (!tud_cdc_connected()) {
        tight_loop_contents();
    }

    printf("hello analog microphone\n");

    // initialize the analog microphone
    if (analog_microphone_init(&config) < 0) {
        printf("analog microphone initialization failed!\n");
        while (1) { tight_loop_contents(); }
    }

    // set callback that is called when all the samples in the library
    // internal sample buffer are ready for reading
    analog_microphone_set_samples_ready_handler(on_analog_samples_ready);
    
    // start capturing data from the analog microphone
    if (analog_microphone_start() < 0) {
        printf("PDM microphone start failed!\n");
        while (1) { tight_loop_contents();  }
    }

    printf("record audio\n");
    while (1) {
        // store and clear the samples read from the callback
        int sample_count = samples_read;
        samples_read = 0;
        
        // loop through any new collected samples
        for (int i = 0; i < sample_count; i++) {
            // printf("%d\n", sample_buffer[i]);
            full_record[full_record_INDEX++] = sample_buffer[i];
            if(full_record_INDEX >= AUDIO_SAMPLE_RATE*RECORD_LENGTH_S){
                break_while = true;
                break;      // break inner for loop in case this does not happen at end of loop
                            // at this point AUDIO_SAMPLE_RATE*RECORD_LENGTH_S number of sample should be available in memory
            }
        }
        if(break_while == true) break;      // break the while() loop if this happens
    }
    printf("Recording complete, stop mic\n");
    analog_microphone_stop();
    analog_microphone_deinit();

    printf("Sending recorded data to serial\n");
    // send the AUDIO_SAMPLE_RATE*RECORD_LENGTH_S number of audio samples to USB serial
    for (int i = 0; i < AUDIO_SAMPLE_RATE*RECORD_LENGTH_S; i++) {
        printf("%d: %d\n",i, full_record[i]);
        sleep_ms(2);
    }

    printf("dummy loop --- \n");
    while(1){

    }

    return 0;
}

I am using a different analog microphone, not the one suggested in this example. Note that I have changed .bias_voltage = 0.0,

I am recording 10 seconds of audio at 8000 Hz entirely in memory, then sending all the recorded audio samples to USB serial port, where I display all the data in putty as well as save as putty log file. I use this log file to regenerate audio in the PC.

If I play back the audio samples in matlab the sound is very fast (only high speed, not high pitch), and sounds as though some audio samples are being skipped. Since the number of audio samples matches what I expect, it seems the skipping is happening at the ADC itself. Could you please give me some hints on how to fix this?

I have uploaded a sample of my recording here: https://soundcloud.com/user-202091011/pico-mic-rec-01-samsung-galaxy-s7-over-the-horizon-2016

The sound volume is a bit low but my main problem is the speed of the playback and the audio frames being skipped

The original sound is Samsung galaxy s7 over the horizon (2016): https://www.youtube.com/watch?v=SyJyB2niF3s

So Actually I'm trying to create a "Sound card" for my Electric Guitar but it didn't work with this Elektret microphone aka. Sound sensor? isn't it compatible or do I need do something I would use PDM but this sort of microphone modules are much easier to replace or solder on it I don't get any soundwave on audacity just a straight line

Hi, I tried to change the sample rate without success (tested only in windows 10). I only changed tusb_config.h to: #define CFG_TUD_AUDIO_EP_SZ_IN (48+1) * CFG_TUD_AUDIO_FUNC_1_N_BYTES_PER_SAMPLE_TX * CFG_TUD_AUDIO_FUNC_1_N_CHANNELS_TX // 16 Samples (16 kHz) x 2 Bytes/Sample x 1 Channel and usb_microphone.h : #define SAMPLE_RATE 48000

any ideas what else need to be modified?

Hi, thank you for the sample code. Microphone sound capture works, but there it comes with background low-frequency tone that fluctuates (on/off a few times per second). I've attached a WAV recording - the fluctuating background tone is easy to hear at the end of the recording. Also, it appears that the audio has some breaks in it (i.e. sound gets missing momentarily). untitled.wav.zip Any help would be much appreciated. Thank you

Hi,

When building the project I get a lot of errors related to TinyUSB -- see below.

Version 1.2 of the Pico SDK updated TinyUSB from 0.8.0 to 0.10.1, and the release notes mention: "Note also that moving from TinyUSB 0.8.0 to TinyUSB 0.10.1 may require some minor changes to your USB code."

I'm guessing that these errors are caused by the TinyUSB update, and changes are required tusb_config.h and usb_descriptors.c -- perhaps main.c as well. I noticed that the TinyUSB audio_test example has quite a few changes to these files.

Can someone please confirm that this project isn't compatible with TinyUSB 0.10.1?

Thanks,

Dan

Errors from build of usb_microphone:

[ 69%] Building C object examples/usb_microphone/CMakeFiles/usb_microphone.dir/main.c.obj In file included from C:\Users\gigam\Pico\pico-sdk\lib\tinyusb\src/tusb.h:80, from C:\Users\gigam\source\repos\pico\pico-microphone-v1\examples\usb_microphone\usb_microphone.h:11, from C:\Users\gigam\source\repos\pico\pico-microphone-v1\examples\usb_microphone\main.c:17: C:\Users\gigam\Pico\pico-sdk\lib\tinyusb\src/class/audio/audio_device.h:40:2: error: #error You must tell the driver the length of the audio function descriptor including IAD descriptor 40 | #error You must tell the driver the length of the audio function descriptor including IAD descriptor | ^~~~~ In file included from C:\Users\gigam\Pico\pico-sdk\lib\tinyusb\src/tusb.h:80, from C:\Users\gigam\source\repos\pico\pico-microphone-v1\examples\usb_microphone\usb_microphone.h:11, from C:\Users\gigam\source\repos\pico\pico-microphone-v1\examples\usb_microphone\main.c:17: C:\Users\gigam\Pico\pico-sdk\lib\tinyusb\src/class/audio/audio_device.h:55:2: error: #error You must tell the driver the number of Standard AS Interface Descriptors you have defined in the audio function descriptor! 55 | #error You must tell the driver the number of Standard AS Interface Descriptors you have defined in the audio function descriptor! | ^~~~~ C:\Users\gigam\Pico\pico-sdk\lib\tinyusb\src/class/audio/audio_device.h:70:2: error: #error You must define an audio class control request buffer size! 70 | #error You must define an audio class control request buffer size!

Hello,

I would really appreciate if you explained the filtering that is implemented for the PDM microphone. Should I be concerned at all about aliasing when using the PDM microphone for an audio classification task with potential sounds above 8kHz? Thanks!

These changes should resolve #6.

They include:

1. a port of https://github.com/hathach/tinyusb/pull/1058

2. bumping CFG_TUD_AUDIO_EP_SZ_IN by one, as per the comment in: https://github.com/hathach/tinyusb/blob/2eaf99e0aa9c10d62dd8d0a4e765f5941bfeaf98/examples/device/audio_4_channel_mic/src/tusb_config.h#L104

// 48 Samples (48 kHz) x 2 Bytes/Sample x CFG_TUD_AUDIO_N_CHANNELS_TX Channels
// - the Windows driver always needs an extra sample per channel of space more, otherwise it complains... found by trial and error

So far, I've tested on a Windows 10 VM and macOS - they seem ok. @iovsiann it would be great if you can test these changes on Windows and Linux and report your feedback.

Assuming the PDM-mic is capaple of doing this: What is the maximum audio sample rate for the Pi pico 2040 and this library? Are 192KHz or even 384KHz possible?

Kind regards

The output frequency of the clk pin is calculated by the sample rate, and some PDM microphone has a limited frequency range for clock speed (Like MP34Dxxx has a clock rate limit from1 to 3.25 MHz), which causes the sample rate below 16000(Corresponding to the clk frequency of 1.024MHz) to sometimes fail to make the microphone work.