Skip to content

transcribeAudioParts

function transcribeAudioParts(input, transcriptionModel): Promise<UserContent>;

Replace audio file parts with their transcript (a text part) using an AI SDK transcription model. With no model configured, audio parts pass through unchanged — audio-capable models (e.g. Gemini) accept them directly. Non-audio parts (images, documents) are always left untouched.

Parameter Type

input

UserContent

transcriptionModel

TranscriptionModel | undefined

Promise<UserContent>