Convert Speech to Text Asynchronously

Description

Converts an audio file to text asynchronously. The file, located publicly in Google Cloud Storage, can have a length of up to 480 minutes.

Check the job status using Get Conversion Status by Job ID and, once completed, download the result with Get Converted Text by Job ID. See Asynchronous conversion for a step-by-step guide.

Project compatibility

Windows | Cross-platform

Configuration

• Connection ID - The connection established in Integration Service. Access the dropdown menu to choose, add, or manage connections.

• Google Cloud Storage URI - The Google Cloud Storage public file URI. For example, gs://uipath-gcs/my-audio-object-name. This field supports String type input.
• Speech language - The language of the voice expressed as a tag, e.g. en-US. This field supports String type input.
• Audio encoding - The encoding of the audio file, such as mp3. This field supports String type input.
• Sample rate in hertz - Sample rate in hertz of the audio data sent in all recognition audio messages. Valid values are: 8000-48000. 16000 is optimal. This field supports String type input.

Manage Properties

Use the Manage Properties wizard to configure or use any of the object's standard or custom fields. You can select fields to add them to the activity canvas. The added standard or custom fields are available in the Properties panel (in Studio Desktop) or under Show additional properties (in Studio Web).

Additional properties

• Enable word time offset - If True, the result includes a list of words and the start and end time offsets (timestamps) for those words. Default is False. Boolean value.

Output

• Job ID - A unique identifier for the long-running transcription operation. Automatically generated output variable.