Attach Watson Transcription

Description

The Attach Watson Transcription converts the results file received from the IBM Watson Speech-to-Text service in json format, converts it to the desired caption format, and adds it to the media package.

Parameter Table

configuration keys	description	default value	example
transcription-job-id	This is filled out by the transcription service when starting the workflow.	EMPTY	Should always be "${transcriptionJobId}"
target-flavor	The flavor of the caption/transcription file generated. Mandatory.	EMPTY	captions/timedtext
target-tag	The tag to apply to the caption/transcription file generated. Optional.	EMPTY	archive
target-caption-format	The caption format to be generated. Optional. If not entered, the raw resulting file will be attached to the media package.	EMPTY	webvtt

Example

<!-- Attach caption/transcript -->
<operation id="attach-watson-transcription"
  fail-on-error="true"
  exception-handler-workflow="ng-partial-error"
  description="Attach captions/transcription">
  <configurations>
    <!-- This is filled out by the transcription service when starting this workflow so just use this as is -->
    <configuration key="transcription-job-id">${transcriptionJobId}</configuration>
    <configuration key="target-flavor">captions/timedtext</configuration>
    <configuration key="target-tag">archive</configuration>
    <configuration key="target-caption-format">webvtt</configuration>
  </configurations>
</operation>

<!-- Merge caption/transcript to existing publication and republish -->
<operation id="publish-engage"
  fail-on-error="true"
  exception-handler-workflow="ng-partial-error"
  description="Distribute and publish to engage server">
  <configurations>
    <configuration key="download-source-flavors">captions/timedtext</configuration>
    <configuration key="strategy">merge</configuration>
    <configuration key="check-availability">true</configuration>
  </configurations>
</operation>