Yandex Cloud
Search
Contact UsGet started
  • Blog
  • Pricing
  • Documentation
  • All Services
  • System Status
    • Featured
    • Infrastructure & Network
    • Data Platform
    • Containers
    • Developer tools
    • Serverless
    • Security
    • Monitoring & Resources
    • ML & AI
    • Business tools
  • All Solutions
    • By industry
    • By use case
    • Economics and Pricing
    • Security
    • Technical Support
    • Customer Stories
    • Gateway to Russia
    • Cloud for Startups
    • Education and Science
  • Blog
  • Pricing
  • Documentation
Yandex project
© 2025 Yandex.Cloud LLC
All solutions
    • All solutions for SpeechKit
    • Fixing the "audio duration should be less than 30s" error in SpeechKit
    • Resolving the "Invalid audio received" error
    • Enabling automated punctuation in the recognized text

In this article:

  • Issue description
  • Solution
  1. SpeechKit
  2. Enabling automated punctuation in the recognized text

Enabling automated punctuation in the recognized text

Written by
Yandex Cloud
Updated at October 17, 2024
  • Issue description
  • Solution

Issue descriptionIssue description

You need to enable automated punctuation (i.e., the punctuator) in the recognized text.

SolutionSolution

The punctuator is available in the following recognition modes:

  1. In gRPC API, the punctuator is enabled by the TEXT_NORMALIZATION_ENABLED option. This option enables the text normalization mode that mostly adds punctuation marks. Acceptable option values are given in this SpeechKit API specification section.

  2. In the asynchronous recognition REST API, the punctuator functionality is enabled by the config.specification.literature_text option. It enables the text normalization mode that mostly adds punctuation marks. Acceptable option values are given in this SpeechKit API specification section.

  3. In the streaming recognition REST API, the punctuator is enabled by the x-normalize-partials option. If this flag is true, you can get intermediate recognition results (parts of a recognized utterance) in a normalized format: numerals are returned as numbers, the profanity filter is enabled, etc. Acceptable option values are given in this SpeechKit API specification section.

Note

Text normalization is currently unavailable in the synchronous recognition mode.

Was the article helpful?

Previous
Resolving the "Invalid audio received" error
Next
All solutions for VPC
Yandex project
© 2025 Yandex.Cloud LLC