Skip to main content

TranscriptionResponse

Result of transcribing an audio clip with speaker diarization --- used by the opt-in auto-crop step to pick the host's segments. Mirrors the TS transcribe response.

textstring

Full transcribed text of the audio clip.

Example: Welcome back to the show. Today we are talking about...
segments object[]

Diarized speaker segments of the transcript, in chronological order.

  • Array [
  • speakerstring

    Diarized speaker label for this segment.

    Example: speaker_0
    textstring

    Transcribed text spoken in this segment.

    Example: This episode is brought to you by Acme...
    startdoublenullable

    Start time of the segment, in seconds from the beginning of the audio.

    Example: 0
    enddoublenullable

    End time of the segment, in seconds from the beginning of the audio.

    Example: 4.2
  • ]
  • languagestringnullable

    Detected language of the audio, as a language code, when available.

    Example: en
    durationdoublenullable

    Total duration of the transcribed audio, in seconds, when available.

    Example: 182.5
    filenamestring

    Original filename of the uploaded audio clip.

    Example: host-read.mp3
    file_sizeint64

    Size of the uploaded audio file, in bytes.

    Example: 2457600
    TranscriptionResponse
    {
    "text": "Welcome back to the show. Today we are talking about...",
    "segments": [
    {
    "speaker": "speaker_0",
    "text": "This episode is brought to you by Acme...",
    "start": 0,
    "end": 4.2
    }
    ],
    "language": "en",
    "duration": 182.5,
    "filename": "host-read.mp3",
    "file_size": 2457600
    }