Automatic Speech Recognition



Automatic Speech Recognition (ASR) Engine receives speech as input and converts it to text.

It can easily recognize continuous speech and record dictation. It is also called Speech to Text (STT).


Image for the process of converting speech to text through ASR Engine


ASR Engine by ThinQ.AI supports the following features:

Features of ASR Engine
Feature Description
Multi-language Support

Supports Korean and English languages

Customized results

Displays results of the speech analysis in intervals throughout dictation, or all at once after dictation is complete.

Secure connection to the server

Establishes connections to the server using the Transport Layer Security (TLS)-based HTTP/2 protocol for security.


Engine Architecture

All functions of ASR Engine are run on a server. ASR Engine receives PCM data and JSON data as input, converts them to text data, and delivers text output accordingly.


Image for the architecture of ASR Engine


Examples of Use

ASR Engine can be used in various ways during everyday life.


  • Use voice to control devices while driving

When hands-on operation of personal devices is difficult or unsafe, use voice commands to send text messages or enter a destination into your navigation system.

Image that sends messages by voice or set the destination of the navigation  by voice while driving

  • Save call center conversations

Save important conversations as text by utilizing the engine's voice recognition function. Conversations with customers can be stored and archived as text.

Image that saves the contents of a call as a text file

  • Prepare minutes

With the advancement of speech recognition technology, speech recognition can be used to automatically create minutes during important events.

Image that the meeting content is written as text