Explaination

• a realtime audio recorder, it can generate pcm or mp3 buffer of audio data with configurable sample rate and bit depth.

• when recording, it start a web worker thread in the browser to transform the bitDepth and sampleRate, so it's very fast in realtime

• please remember to using this package under https, or it won't work

Usage

import Recorder from 'recorder-realtime';

function initRecorder() {
  if (!Recorder.isRecordingSupported()) {
    this.screenLogger('Recording features are not supported in your browser.')
  } else {
    recorder = new Recorder({
      recordingGain: parseInt(this.recordingGain, 10),
      numberOfChannels: 1, // this is suggested
      wavBitDepth: 16,
      format: 'mp3', // default to pcm, which is larger
      wavSampleRate: 16000,
      streamPages: true // we can get the buffer stream in realtime in the ondataavailable callback
    })
    recorder.onstart = () => {
      console.log('recorder started');
    }
    recorder.onstreamerror = e => {
      this.screenLogger('Error encountered: ' + e.message)
    }
    this.recorder.ondataavailable = data => {
      if (data.command === 'buffer') {
        this.sendBlogToServer(data.buffer); // 片段数据，bufferLength的大小
      } else if (data.command === 'wav') {
        const dataBlob = new Blob([typedArray], { type: 'audio/wav' })
        const url = URL.createObjectURL(dataBlob)
        this.audioSrc = url
      }
    }
    recorder.start();
  }
}

Optional Configuration

• bufferLength - (optional) default to 4096, which means when the ondataavailable function called, the size of response buffer will be 4096

• monitorGain - (optional) set the echo of your recording, the value range is 0-1, default to 0, which means mute the echo

• numberOfChannels - (optional) 1 or 2, 1 is suggested

• recordingGain - (optional) the volume of your voice, the value range is 0-1, default to 1

• wavBitDepth - (optional) the bit length of the recordered buffer, default is 16, you can set it to 8, 16, 24, 32