Table Of Contents

Feedback

User Guide

First time using the AWS CLI? See the User Guide for help getting started.

[ aws . transcribe ]

start-transcription-job

Description

Starts an asynchronous job to transcribe speech to text.

See also: AWS API Documentation

See 'aws help' for descriptions of global parameters.

Synopsis

  start-transcription-job
--transcription-job-name <value>
--language-code <value>
[--media-sample-rate-hertz <value>]
[--media-format <value>]
--media <value>
[--output-bucket-name <value>]
[--output-encryption-kms-key-id <value>]
[--settings <value>]
[--job-execution-settings <value>]
[--content-redaction <value>]
[--cli-input-json <value>]
[--generate-cli-skeleton <value>]

Options

--transcription-job-name (string)

The name of the job. Note that you can't use the strings "." or ".." by themselves as the job name. The name must also be unique within an AWS account.

--language-code (string)

The language code for the language used in the input media file.

Possible values:

  • en-US
  • es-US
  • en-AU
  • fr-CA
  • en-GB
  • de-DE
  • pt-BR
  • fr-FR
  • it-IT
  • ko-KR
  • es-ES
  • en-IN
  • hi-IN
  • ar-SA
  • ru-RU
  • zh-CN
  • nl-NL
  • id-ID
  • ta-IN
  • fa-IR
  • en-IE
  • en-AB
  • en-WL
  • pt-PT
  • te-IN
  • tr-TR
  • de-CH
  • he-IL
  • ms-MY
  • ja-JP
  • ar-AE

--media-sample-rate-hertz (integer)

The sample rate, in Hertz, of the audio track in the input media file.

If you do not specify the media sample rate, Amazon Transcribe determines the sample rate. If you specify the sample rate, it must match the sample rate detected by Amazon Transcribe. In most cases, you should leave the MediaSampleRateHertz field blank and let Amazon Transcribe determine the sample rate.

--media-format (string)

The format of the input media file.

Possible values:

  • mp3
  • mp4
  • wav
  • flac

--media (structure)

An object that describes the input media for a transcription job.

Shorthand Syntax:

MediaFileUri=string

JSON Syntax:

{
  "MediaFileUri": "string"
}

--output-bucket-name (string)

The location where the transcription is stored.

If you set the OutputBucketName , Amazon Transcribe puts the transcript in the specified S3 bucket. When you call the GetTranscriptionJob operation, the operation returns this location in the TranscriptFileUri field. If you enable content redaction, the redacted transcript appears in RedactedTranscriptFileUri . If you enable content redaction and choose to output an unredacted transcript, that transcript's location still appears in the TranscriptFileUri . The S3 bucket must have permissions that allow Amazon Transcribe to put files in the bucket. For more information, see Permissions Required for IAM User Roles .

You can specify an AWS Key Management Service (KMS) key to encrypt the output of your transcription using the OutputEncryptionKMSKeyId parameter. If you don't specify a KMS key, Amazon Transcribe uses the default Amazon S3 key for server-side encryption of transcripts that are placed in your S3 bucket.

If you don't set the OutputBucketName , Amazon Transcribe generates a pre-signed URL, a shareable URL that provides secure access to your transcription, and returns it in the TranscriptFileUri field. Use this URL to download the transcription.

--output-encryption-kms-key-id (string)

The Amazon Resource Name (ARN) of the AWS Key Management Service (KMS) key used to encrypt the output of the transcription job. The user calling the StartTranscriptionJob operation must have permission to use the specified KMS key.

You can use either of the following to identify a KMS key in the current account:

  • KMS Key ID: "1234abcd-12ab-34cd-56ef-1234567890ab"
  • KMS Key Alias: "alias/ExampleAlias"

You can use either of the following to identify a KMS key in the current account or another account:

  • Amazon Resource Name (ARN) of a KMS Key: "arn:aws:kms:region:account ID:key/1234abcd-12ab-34cd-56ef-1234567890ab"
  • ARN of a KMS Key Alias: "arn:aws:kms:region:account ID:alias/ExampleAlias"

If you don't specify an encryption key, the output of the transcription job is encrypted with the default Amazon S3 key (SSE-S3).

If you specify a KMS key to encrypt your output, you must also specify an output location in the OutputBucketName parameter.

--settings (structure)

A Settings object that provides optional settings for a transcription job.

Shorthand Syntax:

VocabularyName=string,ShowSpeakerLabels=boolean,MaxSpeakerLabels=integer,ChannelIdentification=boolean,ShowAlternatives=boolean,MaxAlternatives=integer,VocabularyFilterName=string,VocabularyFilterMethod=string

JSON Syntax:

{
  "VocabularyName": "string",
  "ShowSpeakerLabels": true|false,
  "MaxSpeakerLabels": integer,
  "ChannelIdentification": true|false,
  "ShowAlternatives": true|false,
  "MaxAlternatives": integer,
  "VocabularyFilterName": "string",
  "VocabularyFilterMethod": "remove"|"mask"
}

--job-execution-settings (structure)

Provides information about how a transcription job is executed. Use this field to indicate that the job can be queued for deferred execution if the concurrency limit is reached and there are no slots available to immediately run the job.

Shorthand Syntax:

AllowDeferredExecution=boolean,DataAccessRoleArn=string

JSON Syntax:

{
  "AllowDeferredExecution": true|false,
  "DataAccessRoleArn": "string"
}

--content-redaction (structure)

An object that contains the request parameters for content redaction.

Shorthand Syntax:

RedactionType=string,RedactionOutput=string

JSON Syntax:

{
  "RedactionType": "PII",
  "RedactionOutput": "redacted"|"redacted_and_unredacted"
}

--cli-input-json (string) Performs service operation based on the JSON string provided. The JSON string follows the format provided by --generate-cli-skeleton. If other arguments are provided on the command line, the CLI values will override the JSON-provided values. It is not possible to pass arbitrary binary values using a JSON-provided value as the string will be taken literally.

--generate-cli-skeleton (string) Prints a JSON skeleton to standard output without sending an API request. If provided with no value or the value input, prints a sample input JSON that can be used as an argument for --cli-input-json. If provided with the value output, it validates the command inputs and returns a sample output JSON for that command.

See 'aws help' for descriptions of global parameters.

Output

TranscriptionJob -> (structure)

An object containing details of the asynchronous transcription job.

TranscriptionJobName -> (string)

The name of the transcription job.

TranscriptionJobStatus -> (string)

The status of the transcription job.

LanguageCode -> (string)

The language code for the input speech.

MediaSampleRateHertz -> (integer)

The sample rate, in Hertz, of the audio track in the input media file.

MediaFormat -> (string)

The format of the input media file.

Media -> (structure)

An object that describes the input media for the transcription job.

MediaFileUri -> (string)

The S3 object location of the input media file. The URI must be in the same region as the API endpoint that you are calling. The general form is:

s3://bucket-name/keyprefix/objectkey

For example:

s3://examplebucket/example.mp4

s3://examplebucket/mediadocs/example.mp4

For more information about S3 object names, see Object Keys in the Amazon S3 Developer Guide .

Transcript -> (structure)

An object that describes the output of the transcription job.

TranscriptFileUri -> (string)

The S3 object location of the the transcript.

Use this URI to access the transcript. If you specified an S3 bucket in the OutputBucketName field when you created the job, this is the URI of that bucket. If you chose to store the transcript in Amazon Transcribe, this is a shareable URL that provides secure access to that location.

RedactedTranscriptFileUri -> (string)

The S3 object location of the redacted transcript.

Use this URI to access the redacated transcript. If you specified an S3 bucket in the OutputBucketName field when you created the job, this is the URI of that bucket. If you chose to store the transcript in Amazon Transcribe, this is a shareable URL that provides secure access to that location.

StartTime -> (timestamp)

A timestamp that shows with the job was started processing.

CreationTime -> (timestamp)

A timestamp that shows when the job was created.

CompletionTime -> (timestamp)

A timestamp that shows when the job was completed.

FailureReason -> (string)

If the TranscriptionJobStatus field is FAILED , this field contains information about why the job failed.

The FailureReason field can contain one of the following values:

  • Unsupported media format - The media format specified in the MediaFormat field of the request isn't valid. See the description of the MediaFormat field for a list of valid values.
  • The media format provided does not match the detected media format - The media format of the audio file doesn't match the format specified in the MediaFormat field in the request. Check the media format of your media file and make sure that the two values match.
  • Invalid sample rate for audio file - The sample rate specified in the MediaSampleRateHertz of the request isn't valid. The sample rate must be between 8000 and 48000 Hertz.
  • The sample rate provided does not match the detected sample rate - The sample rate in the audio file doesn't match the sample rate specified in the MediaSampleRateHertz field in the request. Check the sample rate of your media file and make sure that the two values match.
  • Invalid file size: file size too large - The size of your audio file is larger than Amazon Transcribe can process. For more information, see Limits in the Amazon Transcribe Developer Guide .
  • Invalid number of channels: number of channels too large - Your audio contains more channels than Amazon Transcribe is configured to process. To request additional channels, see Amazon Transcribe Limits in the Amazon Web Services General Reference .

Settings -> (structure)

Optional settings for the transcription job. Use these settings to turn on speaker recognition, to set the maximum number of speakers that should be identified and to specify a custom vocabulary to use when processing the transcription job.

VocabularyName -> (string)

The name of a vocabulary to use when processing the transcription job.

ShowSpeakerLabels -> (boolean)

Determines whether the transcription job uses speaker recognition to identify different speakers in the input audio. Speaker recognition labels individual speakers in the audio file. If you set the ShowSpeakerLabels field to true, you must also set the maximum number of speaker labels MaxSpeakerLabels field.

You can't set both ShowSpeakerLabels and ChannelIdentification in the same request. If you set both, your request returns a BadRequestException .

MaxSpeakerLabels -> (integer)

The maximum number of speakers to identify in the input audio. If there are more speakers in the audio than this number, multiple speakers will be identified as a single speaker. If you specify the MaxSpeakerLabels field, you must set the ShowSpeakerLabels field to true.

ChannelIdentification -> (boolean)

Instructs Amazon Transcribe to process each audio channel separately and then merge the transcription output of each channel into a single transcription.

Amazon Transcribe also produces a transcription of each item detected on an audio channel, including the start time and end time of the item and alternative transcriptions of the item including the confidence that Amazon Transcribe has in the transcription.

You can't set both ShowSpeakerLabels and ChannelIdentification in the same request. If you set both, your request returns a BadRequestException .

ShowAlternatives -> (boolean)

Determines whether the transcription contains alternative transcriptions. If you set the ShowAlternatives field to true, you must also set the maximum number of alternatives to return in the MaxAlternatives field.

MaxAlternatives -> (integer)

The number of alternative transcriptions that the service should return. If you specify the MaxAlternatives field, you must set the ShowAlternatives field to true.

VocabularyFilterName -> (string)

The name of the vocabulary filter to use when transcribing the audio. The filter that you specify must have the same language code as the transcription job.

VocabularyFilterMethod -> (string)

Set to mask to remove filtered text from the transcript and replace it with three asterisks ("***") as placeholder text. Set to remove to remove filtered text from the transcript without using placeholder text.

JobExecutionSettings -> (structure)

Provides information about how a transcription job is executed.

AllowDeferredExecution -> (boolean)

Indicates whether a job should be queued by Amazon Transcribe when the concurrent execution limit is exceeded. When the AllowDeferredExecution field is true, jobs are queued and will be executed when the number of executing jobs falls below the concurrent execution limit. If the field is false, Amazon Transcribe returns a LimitExceededException exception.

If you specify the AllowDeferredExecution field, you must specify the DataAccessRoleArn field.

DataAccessRoleArn -> (string)

The Amazon Resource Name (ARN) of a role that has access to the S3 bucket that contains the input files. Amazon Transcribe will assume this role to read queued media files. If you have specified an output S3 bucket for the transcription results, this role should have access to the output bucket as well.

If you specify the AllowDeferredExecution field, you must specify the DataAccessRoleArn field.

ContentRedaction -> (structure)

An object that describes content redaction settings for the transcription job.

RedactionType -> (string)

Request parameter that defines the entities to be redacted. The only accepted value is PII .

RedactionOutput -> (string)

Request parameter where you choose whether to output only the redacted transcript or generate an additional unredacted transcript.

When you choose redacted Amazon Transcribe outputs a JSON file with only the redacted transcript and related information.

When you choose redacted_and_unredacted Amazon Transcribe outputs a JSON file with the unredacted transcript and related information in addition to the JSON file with the redacted transcript.