[../deps/mpg123/src/output/coreaudio.c:81] warning: Didn't have any audio data in callback (buffer underflow)

[LayMuiToh]

I got the above failure error message
the code snippet:

      audioCtx = new AudioContext,
      Speaker = require('speaker');
audioCtx.outStream = new Speaker({
  channels: audioCtx.format.numberOfChannels,
  bitDepth: audioCtx.format.bitDepth,
  sampleRate: audioCtx.sampleRate
})

let data;
let count;
let startTime;

var WebSocket = require('ws');
var wsUri = "ws://" + xlVAServer[0].wsURI;
console.log("Connecting to " + wsUri);
var ws = new WebSocket(wsUri);
ws.onopen = function(evt) { onOpen(evt) };
ws.onclose = function(evt) { onClose(evt) };
ws.onmessage = function(evt) { onMessage(evt) };
ws.onerror = function(evt) { onError(evt) };

// Event Handlers to websocket connection
function onOpen(evt) {
  console.log("Connected to server");
}

function onClose(evt) {
  console.log("Disconnected from server");
}

function onError(evt) {
  console.log("Communication error");
}

function onMessage(evt) {
  console.log("Receiving data on websocket ...");
  
  if (evt.data instanceof ArrayBuffer) 
  {
    if (count == 0) {
      startTime = audioCtx.currentTime;
    }
    
    audioCtx.decodeAudioData(evt.data, function(data) {
      count++;
      startTime = startTime + 0.3;
      let duration = playSound(data, startTime);
      console.log('Duration: ' + duration);
      startTime = startTime + duration;
    });
    
  }
}

function playSound(buffer, playTime)
{
  let source = audioCtx.createBufferSource();
  console.log('Audio playback...');
  source.buffer = buffer;
  source.start(playTime);
 
  source.connect(audioCtx.destination);
  console.log('connect destination ');
  let duration = buffer.duration;
  return duration;
}

function perform_tts(text)
{
  data = new ArrayBuffer(0);
  count = 0;
  ws.binaryType = "arraybuffer";
  console.log('Sending text ' + text);
  ws.send(text);
}


export async function playAudio(text, image_name, start_delay, end_delay) {
  switch(xlVAServer[0].useTL_TTS) {
    case 'yes':
      return perform_tts(text);

[LayMuiToh]

and a failure message:
the 2 AudioBuffers don't have the same sampleRate

I don't quite understand why there is 2 audiobuffers here?

[swordsreversed]

@LayMuiToh did you ever figure this out? Having a similar problem.

[dy]

All native dependencies (speaker, mpg123, coreaudio) have been removed. v1.0.0 is pure JS — no native compilation, no system audio dependencies. Real-time output uses speaker package as optional peer dep, or pipe to stdout:

node synth.js | aplay -f cd