split large audio file [help]

[unixxxx]

I have successfully used this library for concatenating audio files, but I was wandering if it can be used for splitting files too.

  const reader = new FileReader();
  const blob = file.slice(offset, chunkSize + offset);
  r.onload = readEventHandler;
  r.readAsArrayBuffer(blob);

reading the file in chunks, and then trying to get AudioBuffer in the readEventHandler

const readEventHandler = evt => {
    const arrayBuffer = evt.target.result as ArrayBuffer;
    crunker._context.decodeAudioData(
          arrayBuffer,
          d => console.log(d),
          e => console.log(e)
    );
}

the problem is that only the first part of the file is getting decoded since only it has the necessary headers. For the rest of the parts I am getting DOMException: Unable to decode audio data.

I tried to manually add the headers to each chunk with modified writeHeaders function , but the audio I get is somehow too fast.

Any suggestions?

const writeHeaders = (arrayBuffer: ArrayBuffer) => {
  // const arrayBuffer = new ArrayBuffer(44 + buffer.length * 2);
  const view = new DataView(arrayBuffer);
  const sampleRate = 44100;

  writeString(view, 0, 'RIFF');
  view.setUint32(4, 32 + arrayBuffer.byteLength * 2, true);
  writeString(view, 8, 'WAVE');
  writeString(view, 12, 'fmt ');
  view.setUint32(16, 16, true);
  view.setUint16(20, 1, true);
  view.setUint16(22, 2, true);
  view.setUint32(24, sampleRate, true);
  view.setUint32(28, sampleRate * 4, true);
  view.setUint16(32, 4, true);
  view.setUint16(34, 16, true);
  writeString(view, 36, 'data');
  view.setUint32(40, arrayBuffer.byteLength * 2, true);

  return floatTo16BitPCM(view, arrayBuffer, 44);
};

[jaggad]

@unixxxx If the audio is too fast, it could be that the arrayBuffer is using the original byteLength rather than the new split length?

Did you have any luck with this? I think splitting would be a good method to add to this project.

[guest271314]

If you use decodeAudioData() first you can get any part of the AudioBuffer from any offset.

If you do not split the file first you can use AudioWorklet or Media Capture Transform to get the exact time slices expected as Float32Arrays and then merge the data.

For live-streaming you can utilize RTCRtpSender.replaceTrack() with MediaRecorder to dynamically change the track(s).

[guest271314]

This https://github.com/guest271314/audioInputToWav (is not fast) will create a WAV file from an AudioBuffer, or when modified, Float32Array(s).

If you are trying to concantenate specific time slices of of multiple audio (and, or, video) Media Fragment URI can be useful https://github.com/guest271314/MediaFragmentRecorder.

[guest271314]

decodeAudioData() requires metadata. It is possible to create files from the output when the entire file is passed to the function.