Streaming Audio while recording it RRS feed

  • Question

  • So in the documentation of Microsoft Speech Recognition API it says, that you can send audio while you are recording it by using Chunked Encoding. I am trying to do this by accessing the REST API with JavaScript. What audio-format am I supposed to use for that? I was trying to do it in .wav, but that obviously didn't work since you need the length of the file to create the header of the .wav file. I also didn't find anything about that in any of the code examples, is this feature implemented yet?

    EDIT, 24.11.16:

    So I figured out that it is possible with .wav-files, by just assuming a length of 0xFFFFFFFF. I am able to create a .wav-file while talking now, however I still don't know how to send a streamed request or chunked requests that are connected.

    • Edited by Selu95 Thursday, November 24, 2016 4:21 PM Made Progress
    Friday, November 18, 2016 5:10 PM