And don't forget to define the master clock pin: w/o this it will not work!
Can you share your pin setup ?

yep. The audiotools sketch you linked to works fine.

It's getting very very close to working. I have some recognizable sound now.

My apologies for writing AudioBoardESP32.h, I meant AudioConfigESP32.h

// Set output mode
//#define ESP32_AUDIO_OUT_MODE INTERNAL_DAC
#define ESP32_AUDIO_OUT_MODE I2S_DAC_AND_I2S_ADC

// For external I2S output, only: I2S_PINS
#define ESP32_I2S_BCK_PIN 27
#define ESP32_I2S_WS_PIN 25
#define ESP32_I2S_DATA_PIN 26
#define ESP32_I2S_DATA_PIN_IN 35

Not sure where to define the master clock pin. The only reference I can find to it is to the old Arduino ADF/AudioKit HAL

I renamed all the existing defines to match the AudioConfigESP32.h names and I'm getting something that resembles the synth sounds I'm used to with the 5102 but it's choppy. Tried changing the sample rate but no change in performance.

the following is the setup called by the sketch with the changes made, but still no master clock, as I don't know where to put it, but it's getting very close! Should I delete any of this that might be conflicting with AudioBoard.h? SAMPLE_RATE is 44100, I tried 44000 as well

void i2sInit() {
i2s_config_t i2s_config = {
.mode = (i2s_mode_t)(I2S_MODE_MASTER | I2S_MODE_TX | I2S_MODE_RX), ////// I added the RX
.sample_rate = SAMPLE_RATE,
.bits_per_sample = I2S_BITS_PER_SAMPLE_16BIT,
.channel_format = I2S_CHANNEL_FMT_RIGHT_LEFT,
.communication_format = (i2s_comm_format_t)(I2S_COMM_FORMAT_STAND_I2S ),
.intr_alloc_flags = ESP_INTR_FLAG_LEVEL2,
.dma_buf_count = DMA_NUM_BUF,
.dma_buf_len = DMA_BUF_LEN,
.use_apll = true,
};

i2s_pin_config_t i2s_pin_config = {
.bck_io_num = ESP32_I2S_BCK_PIN,
.ws_io_num = ESP32_I2S_WS_PIN,
.data_out_num = ESP32_I2S_DATA_PIN,
.data_in_num = ESP32_I2S_DATA_PIN_IN
};

i2s_driver_install(i2s_num, &i2s_config, 0, NULL);

i2s_set_pin(i2s_num, &i2s_pin_config);
i2s_zero_dma_buffer(i2s_num);

// DEBF("I2S is started: BCK %d, WCK %d, DAT %d\r\n", I2S_BCLK_PIN, I2S_WCLK_PIN, I2S_DOUT_PIN);
}

I tried this but it's not changing anything:

In config.h:

#define ESP32_I2S_BCK_PIN 27 // I2S BIT CLOCK pin (BCL BCK CLK)
#define ESP32_I2S_WS_PIN 25 // I2S WORD CLOCK pin (WCK WCL LCK)
#define ESP32_I2S_DATA_PIN 26 // to I2S DATA IN pin (DIN D DAT)
#define ESP32_I2S_DATA_PIN_IN 35
#define ESP32_I2S_MCLK_PIN 0

In setup routine:

i2s_pin_config_t i2s_pin_config = {
.mck_io_num = ESP32_I2S_MCLK_PIN,
.bck_io_num = ESP32_I2S_BCK_PIN,
.ws_io_num = ESP32_I2S_WS_PIN,
.data_out_num = ESP32_I2S_DATA_PIN,
.data_in_num = ESP32_I2S_DATA_PIN_IN
};

Did you confirm with the sine test sketch that these are the correct pins ?
Not sure if the sequency is important, but mck_io_num should usually be last...

I think if you set the log level to debug the logger shows the pins that are effectively used.

AudioKitEs8388V2.begin(cfg); pins are

    // add i2s pins: mclk, bck, ws,data_out, data_in ,(port)
    addI2S(PinFunction::CODEC, 0, 5, 25, 26, 35);

Which does not match with your pins!

So I am asking again which version did you manage to have working with my test sketch (AudioKitEs8388V1, AudioKitEs8388V2 or AudioKitAC101) ?

I just now figured it out. It was the V2 that worked, I tried the mck_io_num last and got errors, it would only compile if I entered it first. It was actually working many hours ago as it turns out. The choppiness in the audio had to do with something in the way the sequencer timing worked . As soon as I commented out /read_button(&buttons[i]);/ in the sequencer, the audio worked perfectly. Strange that it worked fine on a WROOM with no PSRAM but caused a problem with the A1S. Thank you so much for your time. I never would have figured it out without your help. Your work is so vast, it would have taken a month of study just to get an idea.