AI - Speech to Text

** AVAILABLE MARCH / APRIL 2025 **

Overview

The 'AI - Speech to Text' utility action for Power Automate converts speech contained within the file provided to text. 

Input File Types

The following file types are supported:

• FLAC
• MP3
• OPUS
• OGGS
• WAV

Credit Count

Credits are calculated as follows:

1 Minutes = 0.2 credit, for example a 8 minute audio file would equate to 1.6 credits

Subscription Availability

The 'AI - Speech to Text' flow action is available in all Power Automate regions and paid Encodian subscription plans, except the 'Dedicated' plan.  

Parameters

The default 'Utility - Speech to Text' utility action parameters are detailed below:

• File Content: The file content of the source file

Advanced Parameters

The advanced 'Utility - Speech to Text' utility action parameters are detailed below:

• Locale:Set the locale used to transcribe the audio data provided. The default is 'en-US'
• Punctuation Mode: Specify how to handle punctuation
  • None: Disable punctuation 
  • Dictated: Imply explicit (spoken) punctuation
  • Automatic: Instruct the decoder to automatically determine
  • DictatedAndAutomatic: (Default) Use dictated and automatic punctuation
• Profanity Filter Mode: Specify how to handle profanity
  • None: No filtering applied
  • Masked: Replace profanity with asterisks 
  • Removed: Remove profanity from the result
• Speaker Identification: Set how you would like to identify speakers in the conversation
  
  • None: Text will be returned without speaker identification
  • JSON: Text will be returned as a collection of key/value pairs
  • Simplified: Text will be returned as a single text value containing speaker identification marks
• Number of Speakers: Set the maximum number of speakers for speaker identification. Must be less than 36 and larger than or equal to 2

Return Parameters

Action Specific Return Values

• Result - The speech as a string/text value

Standard Return Values

• OperationId - The unique ID assigned to this operation.
• HttpStatusCode - The HTTP Status code for the response.
• HttpStatusMessage - The HTTP Status message for the response.
• Errors - An array of error messages should an error occur.
• Operation Status - Indicates whether the operation has been completed, has been queued or has failed.