Cheeze/src/infra/Audio.cpp

#include "infra/Audio.h"

#include <base64.h>
#include "config/config.h"

#define PRECEDING_BUFFER_SIZE 4

Audio::Audio(MicType micType) {
  _wavData = new char*[_wavDataSize/_dividedWavDataSize];
  for (int i = 0; i < _wavDataSize/_dividedWavDataSize; ++i) 
    _wavData[i] = new char[_dividedWavDataSize];

  _precedingWavBuffer = new CircularBuffer<char*>(PRECEDING_BUFFER_SIZE); // _precedingWavDataSize/_dividedWavDataSize

  i2s = new I2S(micType);
}

Audio::~Audio() {
  for (int i = 0; i < _wavDataSize/_dividedWavDataSize; ++i) delete[] _wavData[i];
  delete[] _wavData;
  delete i2s;

  emptyPrecedingBuffer();
}

void Audio::_createWavHeader(byte* header, int waveDataSize){
  header[0] = 'R';
  header[1] = 'I';
  header[2] = 'F';
  header[3] = 'F';
  unsigned int fileSizeMinus8 = waveDataSize + 44 - 8;
  header[4] = (byte)(fileSizeMinus8 & 0xFF);
  header[5] = (byte)((fileSizeMinus8 >> 8) & 0xFF);
  header[6] = (byte)((fileSizeMinus8 >> 16) & 0xFF);
  header[7] = (byte)((fileSizeMinus8 >> 24) & 0xFF);
  header[8] = 'W';
  header[9] = 'A';
  header[10] = 'V';
  header[11] = 'E';
  header[12] = 'f';
  header[13] = 'm';
  header[14] = 't';
  header[15] = ' ';
  header[16] = 0x10;  // linear PCM
  header[17] = 0x00;
  header[18] = 0x00;
  header[19] = 0x00;
  header[20] = 0x01;  // linear PCM
  header[21] = 0x00;
  header[22] = 0x01;  // monoral
  header[23] = 0x00;
  header[24] = 0x80;  // sampling rate 16000
  header[25] = 0x3E;
  header[26] = 0x00;
  header[27] = 0x00;
  header[28] = 0x00;  // Byte/sec = 16000x2x1 = 32000
  header[29] = 0x7D;
  header[30] = 0x00;
  header[31] = 0x00;
  header[32] = 0x02;  // 16bit monoral
  header[33] = 0x00;
  header[34] = 0x10;  // 16bit
  header[35] = 0x00;
  header[36] = 'd';
  header[37] = 'a';
  header[38] = 't';
  header[39] = 'a';
  header[40] = (byte)(waveDataSize & 0xFF);
  header[41] = (byte)((waveDataSize >> 8) & 0xFF);
  header[42] = (byte)((waveDataSize >> 16) & 0xFF);
  header[43] = (byte)((waveDataSize >> 24) & 0xFF);
}

char** Audio::record() {
  _noiseDetectedInPrecedingBuffer = false;

  _createWavHeader(_paddedHeader, _wavDataSize);

  _startMicros = micros();
  //_readFromPrecedingBuffer(_wavData, _precedingWavDataSize);
  //_recordToBuffer(_wavData + _precedingWavDataSize, _wavDataSize - _precedingWavDataSize);
  _recordToBuffer(_wavData, _wavDataSize);

  return _wavData;
}

void Audio::_recordToBuffer(char** wavBuffer, int waveDataSize) {
  int bitBitPerSample = i2s->GetBitPerSample();

  if (bitBitPerSample == 16) {
    for (int j = 0; j < waveDataSize / _dividedWavDataSize; ++j) {
      i2s->Read(_i2sBuffer, _i2sBufferSize/2);
      _convertBufferToWav(wavBuffer[j]);
    }
  }
  else if (bitBitPerSample == 32) {
    for (int j = 0; j < waveDataSize / _dividedWavDataSize; ++j) {
      i2s->Read(_i2sBuffer, _i2sBufferSize);
      for (int i = 0; i < _i2sBufferSize/8; ++i) {
        wavBuffer[j][2*i] = _i2sBuffer[8*i + 2] << 4; // TODO: check << 4 in 16bit
        wavBuffer[j][2*i + 1] = _i2sBuffer[8*i + 3] << 4;
      }
    }
  }
}

void Audio::emptyPrecedingBuffer() {
  while (!_precedingWavBuffer->isEmpty()) {
    char* tmp;
    _precedingWavBuffer->popFront(tmp);
    delete tmp;
  }
}

void Audio::writeToPrecedingBuffer() {
  if (_precedingWavBuffer->isFull()) {
    char* tmp;
    _precedingWavBuffer->popFront(tmp);
    delete tmp;
  }

  i2s->Read(_i2sBuffer, _i2sBufferSize/2);
  _noiseDetectedInPrecedingBuffer = _isNoiseDetectedInPrecedingBuffer();

  char* _partialWavBuffer = new char[_precedingWavDataSize];
  _convertBufferToWav(_partialWavBuffer);

  CircularBufferErrorCode err = _precedingWavBuffer->pushBack(_partialWavBuffer);
}

void Audio::_readFromPrecedingBuffer(char** wavBuffer, int waveDataSize) {
  char* tmp;

  for (int i = 0; i < PRECEDING_BUFFER_SIZE; i++) {
    _precedingWavBuffer->popFront(tmp);
    wavBuffer[i] = tmp;
  }

  emptyPrecedingBuffer();
}

void Audio::_convertBufferToWav(char* partialWavBuffer) {
  for (int i = 0; i < _i2sBufferSize/8; ++i) {
    partialWavBuffer[2*i] = _i2sBuffer[4*i + 2];
    partialWavBuffer[2*i + 1] = _i2sBuffer[4*i + 3];
  }
}

int Audio::getSize() {
  return (_wavDataSize + sizeof(_paddedHeader)) * 4 / 3; // 4/3 is from base64 encoding
}

String Audio::print(Stream* stream) {
  String enc = base64::encode(_paddedHeader, sizeof(_paddedHeader));
  enc.replace("\n", "");
  stream->print(enc);

  for (int j = 0; j < _wavDataSize / _dividedWavDataSize; ++j) {
    enc = base64::encode((byte*)_wavData[j], _dividedWavDataSize);
    enc.replace("\n", "");
    stream->print(enc);
  }

  return enc;
}

unsigned long Audio::getStartMicros() {
  return _startMicros;
}

bool Audio::_isNoiseDetectedInPrecedingBuffer() {
  if (*(int*)_i2sBuffer > NOISE_THRESHOLD || *(int*)_i2sBuffer < -NOISE_THRESHOLD) {
    Serial.println("Noise detected");
    return true;
  } else {
    return false;
  }
}

bool Audio::isNoiseDetected() {
  return _noiseDetectedInPrecedingBuffer;
}