1.5.2

Changelog:

• transcript create - New parameter --sentiment to include sentiment analysis of recognized utterance.
• transcript create - New parameter --diarization to distinguish between speakers (currently only 2 supported).
• model show - New subcommand.

1.5.1

Maintenance/refactoring update. No NuGet release for this version.

Changelog:

• config create - Removed region check as they appear faster than CLI updates.
• config select - Added check if there is a config with this name in the list.
• model list-scenarios - Includes purpose as well (handy when using "all").
• transcript download - the --out-dir parameter is not required anymore. Current dir will be use if not provided.

Code:

• First attempt at unit testing. Not well yet.
• Code refactored - console app has now its own class.
• Config is now implementing IConfig.

1.5.0

Changelog:

• Added --word-level-timestamps parameter to transcriptions
• (Breaking) Moved transcription ID into argument when downloading transcription (transcript download <ID> instead of transcript download --id <ID>
• model list-scenarios help includes possible values
• Added the ability to upload pronunciation datasets
• Added the ability to create pronunciation models

1.4.1

Version up

1.4.0

New Single transcript command

1.3.2

Fixed: Compile wasn't producing correct archives.

1.3.1

README update to reflect latest changes

1.3.0

List available locales for datasets and models.

1.2.2

User settings moved to profile directory - not tied to CLI location.

1.1.0

Added language models to Test command - required field => breaking change.