How can I tell Actions on Google to stream audio?

According to the documentation you can embed elements in SSML. https://developers.google.com/actions/reference/ssml includes the following example:

<speak>
  Here are <say-as interpet-as="characters">SSML</say-as> samples.
  I can pause <break time="3s"/>.
  I can play a sound
  <audio src="https://www.example.com/MY_MP3_FILE.mp3">didn't get your MP3 audio file</audio>.
  I can speak in cardinals. Your number is <say-as interpret-as="cardinal">10</say-as>.
  Or I can speak in ordinals. You are <say-as interpret-as="ordinal">10</say-as> in line.
  Or I can even speak in digits. The digits for ten are <say-as interpret-as="characters">10</say-as>.
  I can also substitute phrases, like the <sub alias="World Wide Web Consortium">W3C</sub>.
  Finally, I can speak a paragraph with two sentences.
  <p><s>This is sentence one.</s><s>This is sentence two.</s></p>
</speak>

EDIT

p/s : SSML in Documents has these limitations :

• Single channel is preferred, but stereo is acceptable.

• 120 seconds maximum duration. If you want to play audio with a longer duration, consider implementing a media response. 5 megabyte file size limit.

• Source URL must use HTTPS protocol.

• Our UserAgent when fetching the audio is "Google-Speech-Actions".

UPDATE: The first answer works only with the V1 of Dialogflow. As for the V2, you can create the mediaResponse this way (from Google's doc):

conv.ask(new MediaObject({
  name: 'Jazz in Paris',
  url: 'http://storage.googleapis.com/automotive-media/Jazz_In_Paris.mp3',
  description: 'A funky Jazz tune',
  icon: new Image({
    url: 'http://storage.googleapis.com/automotive-media/album_art.jpg',
    alt: 'Media icon',
  }),
}));

========================================================================

I posted an answer over here.

Basically you can create a mediaResponse object that will play your audio file. I can play a 50 mins audio file just fine.

A code example in Node.js could be (with the current documentation):

const richResponse = app.buildRichResponse()
 .addSimpleResponse("Here's song one.")
  .addMediaResponse(app.buildMediaResponse()
  .addMediaObjects([
    app.buildMediaObject("Song One", "https://....mp3")
      .setDescription("Song One with description and large image.") // Optional
      .setImage("https://....jpg", app.Media.ImageType.LARGE)
        // Optional. Use app.Media.ImageType.ICON if displaying icon.
  ])
)
.addSuggestions(["other songs"]);