speech-synthesis-recorder

SpeechSynthesisRecorder.js

Use navigator.mediaDevices.getUserMedia() and MediaRecorder to get audio output from window.speechSynthesis.speak() call as ArrayBuffer, AudioBuffer, Blob, MediaSource, ReadableStream, or other object or data types, see MediaStream, ArrayBuffer, Blob audio result from speak() for recording?.

Usage

Select Monitor of Built-in Audio Analog Stereo option instead of Built-in Audio Analog Stereo option at navigator.mediaDevices.getUserMedia() prompt.

let ttsRecorder = new SpeechSynthesisRecorder({
  text: "The revolution will not be televised", 
  utteranceOptions: {
    voice: "english-us espeak",
    lang: "en-US",
    pitch: .75,
    rate: 1
  }
});

ArrayBuffer

ttsRecorder.start()
  // `tts` : `SpeechSynthesisRecorder` instance, `data` : audio as `dataType` or method call result
  .then(tts => tts.arrayBuffer())
  .then(({tts, data}) => {
    // do stuff with `ArrayBuffer`, `AudioBuffer`, `Blob`,
    // `MediaSource`, `MediaStream`, `ReadableStream`
    // `data` : `ArrayBuffer`
    tts.audioNode.src = URL.createObjectURL(new Blob([data], {type:tts.mimeType}));
    tts.audioNode.title = tts.utterance.text;
    tts.audioNode.onloadedmetadata = () => {
      console.log(tts.audioNode.duration);
      tts.audioNode.play();
    }
  })

AudioBuffer

ttsRecorder.start()
  .then(tts => tts.audioBuffer())
  .then(({tts, data}) => {
    // `data` : `AudioBuffer`
    let source = tts.audioContext.createBufferSource();
    source.buffer = data;
    source.connect(tts.audioContext.destination);
    source.start()
  })

Blob

ttsRecorder.start()
  .then(tts => tts.blob())
  .then(({tts, data}) => {
    // `data` : `Blob`
    tts.audioNode.src = URL.createObjectURL(blob);
    tts.audioNode.title = tts.utterance.text;
    tts.audioNode.onloadedmetadata = () => {
      console.log(tts.audioNode.duration);
      tts.audioNode.play();
    }
  })

ReadableStream

ttsRecorder.start()
  .then(tts => tts.readableStream())
  .then(({tts, data}) => {
    // `data` : `ReadableStream`
    console.log(tts, data);
    data.getReader().read().then(({value, done}) => {
      tts.audioNode.src = URL.createObjectURL(value[0]);
      tts.audioNode.title = tts.utterance.text;
      tts.audioNode.onloadedmetadata = () => {
        console.log(tts.audioNode.duration);
        tts.audioNode.play();
      }
    })
  })

MediaSource

ttsRecorder.start()
  .then(tts => tts.mediaSource())
  .then(({tts, data}) => {
    console.log(tts, data);
    // `data` : `MediaSource`
    tts.audioNode.srcObj = data;
    tts.audioNode.title = tts.utterance.text;
    tts.audioNode.onloadedmetadata = () => {
      console.log(tts.audioNode.duration);
      tts.audioNode.play();
    }
  })

MediaStream

let ttsRecorder = new SpeechSynthesisRecorder({
  text: "The revolution will not be televised", 
  utternanceOptions: {
    voice: "english-us espeak",
    lang: "en-US",
    pitch: .75,
    rate: 1
  }, 
  dataType:"mediaStream"
});
ttsRecorder.start()
  .then(({tts, data}) => {
    // `data` : `MediaStream`
    // do stuff with active `MediaStream`
  })
  .catch(err => console.log(err))

Demo

plnkr