Massive feedback noises

[souravlal]

@marcel-licence Since I was unable to reload samples from SD card I though I could give a try to your new updated code. I have downloaded and uploaded your new code to my ESP32 Audio-kit, but it also doesn't solved my problem. But when ESP32 Audio-kit boots up massive feedback noises are heard and even the noise of me clicking on the buttons can be heard with loud volume. But when I re-upload the old version of your code (without as5600 and es8388 support)
There is not such feedback noises.
By the way the original problem did not get solved either.

[marcel-licence]

#define SAMPLER_ALWAYS_PASS_THROUGH /* can be used to pass line in through audio processing to output */
is set in the config.h

[souravlal]

No, I am still getting the feedback noises with your new code.

[marcel-licence]

Please try to add:
ES8388_SetIn2OoutVOL(0, 0);
Into your setup() routine. That disables the pass through of the ES8388. There are also more possibilities. I highly recommend reading the datasheet.

[souravlal]

Actually I am using the AC101 version with your new code not the es8388 version.

[marcel-licence]

Is the feedback generated when the mainloop is running?
You can put #if 0 ... #endif around the code. There are two options -> the AC101 routes the input signal to the output or the code handles the input and forward it to the output. You can also ensure that input instead of mic input is selected during startup.

[souravlal]

@marcel-licence When the ESP32 Audio-kit starts up the feedback noises are also starts to appear, it continues to do so until I power it off.
Where should I put ' #if 0 ... #endif ' code?
Where should I check whether mic input is selected or not?

[marcel-licence]

ac101_setSourceMic(); <- source is mic, you can also activate line in ac101_setSourceLine
You can set master_output_gain = 0.0f; that should mute the internal processing.
Please look into the code esp32_midi_sampler.ino
/* apply master_output_gain */ fl_sample[n] *= master_output_gain; fr_sample[n] *= master_output_gain;

Following is reading the samples from input:
i2s_read_stereo_samples_buff(fl_sample, fr_sample, SAMPLE_BUFFER_SIZE);

And following is writing samples to output:
if (i2s_write_stereo_samples_buff(fl_sample, fr_sample, SAMPLE_BUFFER_SIZE))