Skip to content

Transcript

Source code in docs_src/capo_transcribe/types/transcript.py
class Transcript(TypedDict, closed=True):
    transcript_file_uri: NotRequired["capo_transcribe.types.uri.Uri"]
    """<p>The Amazon S3 location of your transcript. You can use this URI to access or download your transcript.</p> <p>If you included <code>OutputBucketName</code> in your transcription job request, this is the URI of that bucket. If you also included <code>OutputKey</code> in your request, your output is located in the path you specified in your request.</p> <p>If you didn't include <code>OutputBucketName</code> in your transcription job request, your transcript is stored in a service-managed bucket, and <code>TranscriptFileUri</code> provides you with a temporary URI you can use for secure access to your transcript.</p> <note> <p>Temporary URIs for service-managed Amazon S3 buckets are only valid for 15 minutes. If you get an <code>AccesDenied</code> error, you can get a new temporary URI by running a <code>GetTranscriptionJob</code> or <code>ListTranscriptionJob</code> request.</p> </note>"""
    redacted_transcript_file_uri: NotRequired["capo_transcribe.types.uri.Uri"]
    """<p>The Amazon S3 location of your redacted transcript. You can use this URI to access or download your transcript.</p> <p>If you included <code>OutputBucketName</code> in your transcription job request, this is the URI of that bucket. If you also included <code>OutputKey</code> in your request, your output is located in the path you specified in your request.</p> <p>If you didn't include <code>OutputBucketName</code> in your transcription job request, your transcript is stored in a service-managed bucket, and <code>RedactedTranscriptFileUri</code> provides you with a temporary URI you can use for secure access to your transcript.</p> <note> <p>Temporary URIs for service-managed Amazon S3 buckets are only valid for 15 minutes. If you get an <code>AccesDenied</code> error, you can get a new temporary URI by running a <code>GetTranscriptionJob</code> or <code>ListTranscriptionJob</code> request.</p> </note>"""

redacted_transcript_file_uri instance-attribute

redacted_transcript_file_uri: NotRequired[Uri]

The Amazon S3 location of your redacted transcript. You can use this URI to access or download your transcript.

If you included OutputBucketName in your transcription job request, this is the URI of that bucket. If you also included OutputKey in your request, your output is located in the path you specified in your request.

If you didn't include OutputBucketName in your transcription job request, your transcript is stored in a service-managed bucket, and RedactedTranscriptFileUri provides you with a temporary URI you can use for secure access to your transcript.

Temporary URIs for service-managed Amazon S3 buckets are only valid for 15 minutes. If you get an AccesDenied error, you can get a new temporary URI by running a GetTranscriptionJob or ListTranscriptionJob request.

transcript_file_uri instance-attribute

transcript_file_uri: NotRequired[Uri]

The Amazon S3 location of your transcript. You can use this URI to access or download your transcript.

If you included OutputBucketName in your transcription job request, this is the URI of that bucket. If you also included OutputKey in your request, your output is located in the path you specified in your request.

If you didn't include OutputBucketName in your transcription job request, your transcript is stored in a service-managed bucket, and TranscriptFileUri provides you with a temporary URI you can use for secure access to your transcript.

Temporary URIs for service-managed Amazon S3 buckets are only valid for 15 minutes. If you get an AccesDenied error, you can get a new temporary URI by running a GetTranscriptionJob or ListTranscriptionJob request.