Skip to content

Latest commit

 

History

History
16 lines (13 loc) · 878 Bytes

README.md

File metadata and controls

16 lines (13 loc) · 878 Bytes

SpeechToTextWithAADToken

Using Azure Speech SDK with Azure Active Directory Authentication

This is a simple node application to demostrate how to use Speech SDK for transcription with Azure Active Directory token.

Prerequisites

  1. A Speech Service resource in Azure with custom domain.
  2. A Service Principal for Cognitive Services access permissions

How to run this code

  1. Clone the repository
  2. Create a .env file in SpeechToTextWithAADToken directory and create a variable RESOURCE_ID="" in that file and set your speech resource ID
  3. Copy the wav file you want to transcribe in SpeechToTextWithAADToken directory
  4. When running first time you must run command "npm install" in the terminal
  5. Run the program by running the command "node SpeechRecognition.js"
  6. Once program is finished transcribing it will log the output in the terminal