nova-3

Indicates the configuration category or payload type.

Sets the language to use for processing, transcription, or generation.

Applies automatic formatting improvements for cleaner final output.

Automatically identifies the language of the input or generated text.

Separates and labels different speakers in audio transcription tasks.

Adds or normalizes punctuation in generated or transcribed text.

Controls formatting behavior for numbers in generated or transcribed text.

Controls whether output is organized and split into paragraph blocks.

Stores utterance-level data, often used for speech or dialogue processing.

Controls whether filler words like um and uh are preserved or removed.

Enables detection of named entities such as people, places, and organizations.