• Documentation
    • Search documentation
    karat

    +

    K

    API Reference ↗
    Data connectivity & integrationPipeline Builder ExpressionsTranscribe audio into text

    Transcribe audio into text

    Supported in: Batch

    Transcribe audio files into text.

    Expression categories: Media

    Declared arguments

    • Media reference - The column containing media references to audio files in the media sets.
      Expression<Media reference>
    • optional Language - The language to detect in the input file. If no language is provided, it will be inferred from the first 30 seconds of audio.
      Enum<Afrikaans, Albanian, Amharic, Arabic, Armenian, Assamese, Azerbaijani, Bashkir, Basque, Belarusian, and more ...>
    • optional Output mode - Choose to output as a simple output where the output is the type of the output type parameter and errors are returned as null, or output a struct with the output the output type and error as fields.
      Enum<Simple, With errors>
    • optional Performance mode - The performance mode to use when running transcription. If no mode is provided, we will default to the more economical option.
      Enum<More economical, More performant>

    Output type: String | Struct<ok, error>

    Examples

    Example 1: Base case

    Description: Transcribe the audio file Argument values:

    • Media reference: mediaReference
    • Language: null
    • Output mode: null
    • Performance mode: null
    mediaReferenceOutput
    {"mimeType":"audio/mpeg","reference":{"type":"mediaSetItem","mediaSetItem":{"mediaSetRid":"ri.mio.main.media-set.a", "mediaItemRid":"ri.mio.main.media-item.a"}}}This is an example transcription from Whisper

    Example 2: Null case

    Argument values:

    • Media reference: Media Reference
    • Language: null
    • Output mode: null
    • Performance mode: null
    mediaReferenceOutput
    nullnull