Skip to content
This repository has been archived by the owner on Aug 14, 2019. It is now read-only.

How do you configure the end silence timeout for DataRecognitionClient & MicrophoneRecognitionClient #62

Open
BeanStalka opened this issue Nov 2, 2017 · 2 comments

Comments

@BeanStalka
Copy link

BeanStalka commented Nov 2, 2017

Is there anyway I can configure the amount of milliseconds that ellapse before BING considers an utterace complete. IOW, can I set a end silence timeout.

Example:

1.) User: "Whats the weather like in Philadelphia PA?"
2.) Voiceless audio is captured for 'N' milliseconds.
3.) Bing finalizes its listening/processing and calls into OnFinalResponseReceived

I need to configue 'N' in step number two. So I can wait longer, or shorter amounts of time before declaring an utterance complete.

I'm pretty sure BING is doing this with a static 1000 ms wait after speech has completed. I would like to trim this up to 750 ms. and see how it feels.

I would like to have the method of doing so for both the DataRecognitionClient & MicrophoneRecognitionClient if its at all possible.

FYI, I am using SpeechRecognitionMode.ShortPhrase

Please Advise,

Bean

@wolfma61
Copy link
Contributor

wolfma61 commented Nov 7, 2017

sorry - this is currently not configurable. Hopefully it will be available in a future version
Wolfgang

@radhasoni65
Copy link

hello @wolfma61 ,

Can you please suggest me how to detect when i pause to spoke while recording audio for Speech to Text Conversion by using Bing Speech .

Steps to Produce this:-

  1. Start recording audio .
  2. I spoke 9 seconds after that i just pause for a second than i got the result.
  3. I don't want the result at the time when people take break for a second while speaking.
    So can we manage the Utterance duration while recoding .

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants