AmberScript Start Transcription Workflow Operation
ID: amberscript-start-transcription
Description
The AmberScript Start Transcription operation invokes the AmberScript transcription service by submitting an audio or video file to be transcribed and captioned.
Parameter Table
| configuration keys | description | default | example |
|---|---|---|---|
| source-tag | A tag selecting the audio or video file to be sent for translation/transcription. | - | engage-download |
| source-flavor | A flavor selecting the audio or video file to be sent for translation/transcription. | - | */themed |
| jobtype | direct (automated, fast) or perfect (additional manual improvements, slow). | direct | perfect |
| language | The assumed language for transcription. | en | nl |
| speaker | The number of speakers in the recording. | 1 | 2 |
| transcriptiontype | transcription, captions or translatedSubtitles | transcription | captions |
| glossary | An ID of a custom glossary to use. | - | 643966c3f3e91e0c96e9e060 |
| transcriptionstyle | cleanread or verbatim | cleanread | verbatim |
| targetlanguage | The language of the resulting transcriptions. | - | en |
| skip-if-flavor-exists | If this flavor already exists in the media package, skip this operation. | captions/vtt | captions/timedtext |
Supported Languages
At time of writing Amberscript supports the following language codes:
- da
- de
- en
- es
- fi
- fr
- nl
- no
- sv
Glossaries
The glossary ID needs to be a valid ID for a previously created custom glossary.
See also https://amberscript.github.io/api-docs/#glossary.
If this parameter is unset, the default configured in
etc/org.opencastproject.transcription.amberscript.AmberscriptTranscriptionService.cfg
is used.
If you don't want to use a glossary in a specific workflow (instance)
even though a fallback is configured in that file,
you can provide an empty string as glossary ID, i.e. glossary="".