Skip to main content

AI-Powered Real-Time Speech Recognition

This feature enables real-time transcription of audio content in RTC calls, video conferences, audio chat rooms, and live streaming scenarios. The AI-powered real-time speech recognition delivers high accuracy with low latency.

Currently supports recognition for over 50 languages including Chinese, English, Japanese, Korean, Arabic, French, Spanish, Thai, Indonesian, etc. See Language Code List for details.

Prerequisites​

AI-powered real-time speech recognition is an advanced feature of RC RTC SDK. To enable this feature, please submit a ticket.

Configuring Speech Recognition UI Visibility​

You can control whether to display the speech recognition UI.

API​

Before initiating or answering a call, configure the speech recognition UI visibility via the displayASRUI property of RCCall. Defaults to visible.

tip

To avoid UI display issues, configure this setting before call initiation.

API Prototype​

@interface RCCall : NSObject
/*!
 Controls speech recognition UI visibility. Default: YES (visible). Configure before call to prevent UI issues.
*/
@property (nonatomic, assign) BOOL displayASRUI;
@end

Sample Code​

// Show speech recognition UI
[RCCall sharedRCCall].displayASRUI = YES;
Last updated on