Synchronous audio recognition

Written by

Improved by

Updated at April 30, 2025

Warning

This feature is only available in the Russia region.

Synchronous audio recognition ensures fast response times and is suitable for pre-recorded small single-channel audio fragments.

If you want to recognize speech over the same connection, use streaming mode. In streaming mode, you can get intermediate recognition results.

Audio requirements

The audio you send must meet the following requirements:

If your file is larger, longer, or has more audio channels, use asynchronous recognition.

SpeechKit allows you to recognize and synthesize the following audio formats:

For more information about each format's special features, see Supported audio formats.