StartTranscriptionJob - Amazon Transcribe

StartTranscriptionJob

Transcribes the audio from a media file and applies any additional Request Parameters you choose to include in your request.

To make a StartTranscriptionJob request, you must first upload your media file into an Amazon S3 bucket; you can then specify the Amazon S3 location of the file using the Media parameter.

You must include the following parameters in your StartTranscriptionJob request:

  • region: The AWS Region where you are making your request. For a list of AWS Regions supported with Amazon Transcribe, refer to Amazon Transcribe endpoints and quotas.

  • TranscriptionJobName: A custom name you create for your transcription job that is unique within your AWS account.

  • Media (MediaFileUri): The Amazon S3 location of your media file.

  • One of LanguageCode, IdentifyLanguage, or IdentifyMultipleLanguages: If you know the language of your media file, specify it using the LanguageCode parameter; you can find all valid language codes in the Supported languages table. If you do not know the languages spoken in your media, use either IdentifyLanguage or IdentifyMultipleLanguages and let Amazon Transcribe identify the languages for you.

Request Syntax

{ "ContentRedaction": { "PiiEntityTypes": [ "string" ], "RedactionOutput": "string", "RedactionType": "string" }, "IdentifyLanguage": boolean, "IdentifyMultipleLanguages": boolean, "JobExecutionSettings": { "AllowDeferredExecution": boolean, "DataAccessRoleArn": "string" }, "KMSEncryptionContext": { "string" : "string" }, "LanguageCode": "string", "LanguageIdSettings": { "string" : { "LanguageModelName": "string", "VocabularyFilterName": "string", "VocabularyName": "string" } }, "LanguageOptions": [ "string" ], "Media": { "MediaFileUri": "string", "RedactedMediaFileUri": "string" }, "MediaFormat": "string", "MediaSampleRateHertz": number, "ModelSettings": { "LanguageModelName": "string" }, "OutputBucketName": "string", "OutputEncryptionKMSKeyId": "string", "OutputKey": "string", "Settings": { "ChannelIdentification": boolean, "MaxAlternatives": number, "MaxSpeakerLabels": number, "ShowAlternatives": boolean, "ShowSpeakerLabels": boolean, "VocabularyFilterMethod": "string", "VocabularyFilterName": "string", "VocabularyName": "string" }, "Subtitles": { "Formats": [ "string" ], "OutputStartIndex": number }, "Tags": [ { "Key": "string", "Value": "string" } ], "ToxicityDetection": [ { "ToxicityCategories": [ "string" ] } ], "TranscriptionJobName": "string" }

Request Parameters

For information about the parameters that are common to all actions, see Common Parameters.

The request accepts the following data in JSON format.

ContentRedaction

Makes it possible to redact or flag specified personally identifiable information (PII) in your transcript. If you use ContentRedaction, you must also include the sub-parameters: RedactionOutput and RedactionType. You can optionally include PiiEntityTypes to choose which types of PII you want to redact. If you do not include PiiEntityTypes in your request, all PII is redacted.

Type: ContentRedaction object

Required: No

IdentifyLanguage

Enables automatic language identification in your transcription job request. Use this parameter if your media file contains only one language. If your media contains multiple languages, use IdentifyMultipleLanguages instead.

If you include IdentifyLanguage, you can optionally include a list of language codes, using LanguageOptions, that you think may be present in your media file. Including LanguageOptions restricts IdentifyLanguage to only the language options that you specify, which can improve transcription accuracy.

If you want to apply a custom language model, a custom vocabulary, or a custom vocabulary filter to your automatic language identification request, include LanguageIdSettings with the relevant sub-parameters (VocabularyName, LanguageModelName, and VocabularyFilterName). If you include LanguageIdSettings, also include LanguageOptions.

Note that you must include one of LanguageCode, IdentifyLanguage, or IdentifyMultipleLanguages in your request. If you include more than one of these parameters, your transcription job fails.

Type: Boolean

Required: No

IdentifyMultipleLanguages

Enables automatic multi-language identification in your transcription job request. Use this parameter if your media file contains more than one language. If your media contains only one language, use IdentifyLanguage instead.

If you include IdentifyMultipleLanguages, you can optionally include a list of language codes, using LanguageOptions, that you think may be present in your media file. Including LanguageOptions restricts IdentifyLanguage to only the language options that you specify, which can improve transcription accuracy.

If you want to apply a custom vocabulary or a custom vocabulary filter to your automatic language identification request, include LanguageIdSettings with the relevant sub-parameters (VocabularyName and VocabularyFilterName). If you include LanguageIdSettings, also include LanguageOptions.

Note that you must include one of LanguageCode, IdentifyLanguage, or IdentifyMultipleLanguages in your request. If you include more than one of these parameters, your transcription job fails.

Type: Boolean

Required: No

JobExecutionSettings

Makes it possible to control how your transcription job is processed. Currently, the only JobExecutionSettings modification you can choose is enabling job queueing using the AllowDeferredExecution sub-parameter.

If you include JobExecutionSettings in your request, you must also include the sub-parameters: AllowDeferredExecution and DataAccessRoleArn.

Type: JobExecutionSettings object

Required: No

KMSEncryptionContext

A map of plain text, non-secret key:value pairs, known as encryption context pairs, that provide an added layer of security for your data. For more information, see AWS KMS encryption context and Asymmetric keys in AWS KMS.

Type: String to string map

Map Entries: Maximum number of 10 items.

Key Length Constraints: Minimum length of 1. Maximum length of 2000.

Key Pattern: .*\S.*

Value Length Constraints: Minimum length of 1. Maximum length of 2000.

Value Pattern: .*\S.*

Required: No

LanguageCode

The language code that represents the language spoken in the input media file.

If you're unsure of the language spoken in your media file, consider using IdentifyLanguage or IdentifyMultipleLanguages to enable automatic language identification.

Note that you must include one of LanguageCode, IdentifyLanguage, or IdentifyMultipleLanguages in your request. If you include more than one of these parameters, your transcription job fails.

For a list of supported languages and their associated language codes, refer to the Supported languages table.

Note

To transcribe speech in Modern Standard Arabic (ar-SA), your media file must be encoded at a sample rate of 16,000 Hz or higher.

Type: String

Valid Values: af-ZA | ar-AE | ar-SA | da-DK | de-CH | de-DE | en-AB | en-AU | en-GB | en-IE | en-IN | en-US | en-WL | es-ES | es-US | fa-IR | fr-CA | fr-FR | he-IL | hi-IN | id-ID | it-IT | ja-JP | ko-KR | ms-MY | nl-NL | pt-BR | pt-PT | ru-RU | ta-IN | te-IN | tr-TR | zh-CN | zh-TW | th-TH | en-ZA | en-NZ | vi-VN | sv-SE | ab-GE | ast-ES | az-AZ | ba-RU | be-BY | bg-BG | bn-IN | bs-BA | ca-ES | ckb-IQ | ckb-IR | cs-CZ | cy-WL | el-GR | et-ET | eu-ES | fi-FI | gl-ES | gu-IN | ha-NG | hr-HR | hu-HU | hy-AM | is-IS | ka-GE | kab-DZ | kk-KZ | kn-IN | ky-KG | lg-IN | lt-LT | lv-LV | mhr-RU | mi-NZ | mk-MK | ml-IN | mn-MN | mr-IN | mt-MT | no-NO | or-IN | pa-IN | pl-PL | ps-AF | ro-RO | rw-RW | si-LK | sk-SK | sl-SI | so-SO | sr-RS | su-ID | sw-BI | sw-KE | sw-RW | sw-TZ | sw-UG | tl-PH | tt-RU | ug-CN | uk-UA | uz-UZ | wo-SN | zu-ZA

Required: No

LanguageIdSettings

If using automatic language identification in your request and you want to apply a custom language model, a custom vocabulary, or a custom vocabulary filter, include LanguageIdSettings with the relevant sub-parameters (VocabularyName, LanguageModelName, and VocabularyFilterName). Note that multi-language identification (IdentifyMultipleLanguages) doesn't support custom language models.

LanguageIdSettings supports two to five language codes. Each language code you include can have an associated custom language model, custom vocabulary, and custom vocabulary filter. The language codes that you specify must match the languages of the associated custom language models, custom vocabularies, and custom vocabulary filters.

It's recommended that you include LanguageOptions when using LanguageIdSettings to ensure that the correct language dialect is identified. For example, if you specify a custom vocabulary that is in en-US but Amazon Transcribe determines that the language spoken in your media is en-AU, your custom vocabulary is not applied to your transcription. If you include LanguageOptions and include en-US as the only English language dialect, your custom vocabulary is applied to your transcription.

If you want to include a custom language model with your request but do not want to use automatic language identification, use instead the ModelSettings parameter with the LanguageModelName sub-parameter. If you want to include a custom vocabulary or a custom vocabulary filter (or both) with your request but do not want to use automatic language identification, use instead the Settings parameter with the VocabularyName or VocabularyFilterName (or both) sub-parameter.

Type: String to LanguageIdSettings object map

Map Entries: Maximum number of 5 items.

Valid Keys: af-ZA | ar-AE | ar-SA | da-DK | de-CH | de-DE | en-AB | en-AU | en-GB | en-IE | en-IN | en-US | en-WL | es-ES | es-US | fa-IR | fr-CA | fr-FR | he-IL | hi-IN | id-ID | it-IT | ja-JP | ko-KR | ms-MY | nl-NL | pt-BR | pt-PT | ru-RU | ta-IN | te-IN | tr-TR | zh-CN | zh-TW | th-TH | en-ZA | en-NZ | vi-VN | sv-SE | ab-GE | ast-ES | az-AZ | ba-RU | be-BY | bg-BG | bn-IN | bs-BA | ca-ES | ckb-IQ | ckb-IR | cs-CZ | cy-WL | el-GR | et-ET | eu-ES | fi-FI | gl-ES | gu-IN | ha-NG | hr-HR | hu-HU | hy-AM | is-IS | ka-GE | kab-DZ | kk-KZ | kn-IN | ky-KG | lg-IN | lt-LT | lv-LV | mhr-RU | mi-NZ | mk-MK | ml-IN | mn-MN | mr-IN | mt-MT | no-NO | or-IN | pa-IN | pl-PL | ps-AF | ro-RO | rw-RW | si-LK | sk-SK | sl-SI | so-SO | sr-RS | su-ID | sw-BI | sw-KE | sw-RW | sw-TZ | sw-UG | tl-PH | tt-RU | ug-CN | uk-UA | uz-UZ | wo-SN | zu-ZA

Required: No

LanguageOptions

You can specify two or more language codes that represent the languages you think may be present in your media. Including more than five is not recommended. If you're unsure what languages are present, do not include this parameter.

If you include LanguageOptions in your request, you must also include IdentifyLanguage.

For more information, refer to Supported languages.

To transcribe speech in Modern Standard Arabic (ar-SA), your media file must be encoded at a sample rate of 16,000 Hz or higher.

Type: Array of strings

Array Members: Minimum number of 1 item.

Valid Values: af-ZA | ar-AE | ar-SA | da-DK | de-CH | de-DE | en-AB | en-AU | en-GB | en-IE | en-IN | en-US | en-WL | es-ES | es-US | fa-IR | fr-CA | fr-FR | he-IL | hi-IN | id-ID | it-IT | ja-JP | ko-KR | ms-MY | nl-NL | pt-BR | pt-PT | ru-RU | ta-IN | te-IN | tr-TR | zh-CN | zh-TW | th-TH | en-ZA | en-NZ | vi-VN | sv-SE | ab-GE | ast-ES | az-AZ | ba-RU | be-BY | bg-BG | bn-IN | bs-BA | ca-ES | ckb-IQ | ckb-IR | cs-CZ | cy-WL | el-GR | et-ET | eu-ES | fi-FI | gl-ES | gu-IN | ha-NG | hr-HR | hu-HU | hy-AM | is-IS | ka-GE | kab-DZ | kk-KZ | kn-IN | ky-KG | lg-IN | lt-LT | lv-LV | mhr-RU | mi-NZ | mk-MK | ml-IN | mn-MN | mr-IN | mt-MT | no-NO | or-IN | pa-IN | pl-PL | ps-AF | ro-RO | rw-RW | si-LK | sk-SK | sl-SI | so-SO | sr-RS | su-ID | sw-BI | sw-KE | sw-RW | sw-TZ | sw-UG | tl-PH | tt-RU | ug-CN | uk-UA | uz-UZ | wo-SN | zu-ZA

Required: No

Media

Describes the Amazon S3 location of the media file you want to use in your request.

Type: Media object

Required: Yes

MediaFormat

Specify the format of your input media file.

Type: String

Valid Values: mp3 | mp4 | wav | flac | ogg | amr | webm | m4a

Required: No

MediaSampleRateHertz

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

If you do not specify the media sample rate, Amazon Transcribe determines it for you. If you specify the sample rate, it must match the rate detected by Amazon Transcribe. If there's a mismatch between the value that you specify and the value detected, your job fails. In most cases, you can omit MediaSampleRateHertz and let Amazon Transcribe determine the sample rate.

Type: Integer

Valid Range: Minimum value of 8000. Maximum value of 48000.

Required: No

ModelSettings

Specify the custom language model you want to include with your transcription job. If you include ModelSettings in your request, you must include the LanguageModelName sub-parameter.

For more information, see Custom language models.

Type: ModelSettings object

Required: No

OutputBucketName

The name of the Amazon S3 bucket where you want your transcription output stored. Do not include the S3:// prefix of the specified bucket.

If you want your output to go to a sub-folder of this bucket, specify it using the OutputKey parameter; OutputBucketName only accepts the name of a bucket.

For example, if you want your output stored in S3://DOC-EXAMPLE-BUCKET, set OutputBucketName to DOC-EXAMPLE-BUCKET. However, if you want your output stored in S3://DOC-EXAMPLE-BUCKET/test-files/, set OutputBucketName to DOC-EXAMPLE-BUCKET and OutputKey to test-files/.

Note that Amazon Transcribe must have permission to use the specified location. You can change Amazon S3 permissions using the AWS Management Console. See also Permissions Required for IAM User Roles.

If you do not specify OutputBucketName, your transcript is placed in a service-managed Amazon S3 bucket and you are provided with a URI to access your transcript.

Type: String

Length Constraints: Maximum length of 64.

Pattern: [a-z0-9][\.\-a-z0-9]{1,61}[a-z0-9]

Required: No

OutputEncryptionKMSKeyId

The KMS key you want to use to encrypt your transcription output.

If using a key located in the current AWS account, you can specify your KMS key in one of four ways:

  1. Use the KMS key ID itself. For example, 1234abcd-12ab-34cd-56ef-1234567890ab.

  2. Use an alias for the KMS key ID. For example, alias/ExampleAlias.

  3. Use the Amazon Resource Name (ARN) for the KMS key ID. For example, arn:aws:kms:region:account-ID:key/1234abcd-12ab-34cd-56ef-1234567890ab.

  4. Use the ARN for the KMS key alias. For example, arn:aws:kms:region:account-ID:alias/ExampleAlias.

If using a key located in a different AWS account than the current AWS account, you can specify your KMS key in one of two ways:

  1. Use the ARN for the KMS key ID. For example, arn:aws:kms:region:account-ID:key/1234abcd-12ab-34cd-56ef-1234567890ab.

  2. Use the ARN for the KMS key alias. For example, arn:aws:kms:region:account-ID:alias/ExampleAlias.

If you do not specify an encryption key, your output 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 using the OutputLocation parameter.

Note that the role making the StartCallAnalyticsJob request must have permission to use the specified KMS key.

Type: String

Length Constraints: Minimum length of 1. Maximum length of 2048.

Pattern: ^[A-Za-z0-9][A-Za-z0-9:_/+=,@.-]{0,2048}$

Required: No

OutputKey

Use in combination with OutputBucketName to specify the output location of your transcript and, optionally, a unique name for your output file. The default name for your transcription output is the same as the name you specified for your transcription job (TranscriptionJobName).

Here are some examples of how you can use OutputKey:

  • If you specify 'DOC-EXAMPLE-BUCKET' as the OutputBucketName and 'my-transcript.json' as the OutputKey, your transcription output path is s3://DOC-EXAMPLE-BUCKET/my-transcript.json.

  • If you specify 'my-first-transcription' as the TranscriptionJobName, 'DOC-EXAMPLE-BUCKET' as the OutputBucketName, and 'my-transcript' as the OutputKey, your transcription output path is s3://DOC-EXAMPLE-BUCKET/my-transcript/my-first-transcription.json.

  • If you specify 'DOC-EXAMPLE-BUCKET' as the OutputBucketName and 'test-files/my-transcript.json' as the OutputKey, your transcription output path is s3://DOC-EXAMPLE-BUCKET/test-files/my-transcript.json.

  • If you specify 'my-first-transcription' as the TranscriptionJobName, 'DOC-EXAMPLE-BUCKET' as the OutputBucketName, and 'test-files/my-transcript' as the OutputKey, your transcription output path is s3://DOC-EXAMPLE-BUCKET/test-files/my-transcript/my-first-transcription.json.

If you specify the name of an Amazon S3 bucket sub-folder that doesn't exist, one is created for you.

Type: String

Length Constraints: Minimum length of 1. Maximum length of 1024.

Pattern: [a-zA-Z0-9-_.!*'()/]{1,1024}$

Required: No

Settings

Specify additional optional settings in your StartTranscriptionJob request, including channel identification, alternative transcriptions, speaker partitioning. You can use that to apply custom vocabularies and vocabulary filters.

If you want to include a custom vocabulary or a custom vocabulary filter (or both) with your request but do not want to use automatic language identification, use Settings with the VocabularyName or VocabularyFilterName (or both) sub-parameter.

If you're using automatic language identification with your request and want to include a custom language model, a custom vocabulary, or a custom vocabulary filter, use instead the LanguageIdSettings parameter with the LanguageModelName, VocabularyName or VocabularyFilterName sub-parameters.

Type: Settings object

Required: No

Subtitles

Produces subtitle files for your input media. You can specify WebVTT (*.vtt) and SubRip (*.srt) formats.

Type: Subtitles object

Required: No

Tags

Adds one or more custom tags, each in the form of a key:value pair, to a new transcription job at the time you start this new job.

To learn more about using tags with Amazon Transcribe, refer to Tagging resources.

Type: Array of Tag objects

Array Members: Minimum number of 1 item. Maximum number of 200 items.

Required: No

ToxicityDetection

Enables toxic speech detection in your transcript. If you include ToxicityDetection in your request, you must also include ToxicityCategories.

For information on the types of toxic speech Amazon Transcribe can detect, see Detecting toxic speech.

Type: Array of ToxicityDetectionSettings objects

Array Members: Fixed number of 1 item.

Required: No

TranscriptionJobName

A unique name, chosen by you, for your transcription job. The name that you specify is also used as the default name of your transcription output file. If you want to specify a different name for your transcription output, use the OutputKey parameter.

This name is case sensitive, cannot contain spaces, and must be unique within an AWS account. If you try to create a new job with the same name as an existing job, you get a ConflictException error.

Type: String

Length Constraints: Minimum length of 1. Maximum length of 200.

Pattern: ^[0-9a-zA-Z._-]+

Required: Yes

Response Syntax

{ "TranscriptionJob": { "CompletionTime": number, "ContentRedaction": { "PiiEntityTypes": [ "string" ], "RedactionOutput": "string", "RedactionType": "string" }, "CreationTime": number, "FailureReason": "string", "IdentifiedLanguageScore": number, "IdentifyLanguage": boolean, "IdentifyMultipleLanguages": boolean, "JobExecutionSettings": { "AllowDeferredExecution": boolean, "DataAccessRoleArn": "string" }, "LanguageCode": "string", "LanguageCodes": [ { "DurationInSeconds": number, "LanguageCode": "string" } ], "LanguageIdSettings": { "string" : { "LanguageModelName": "string", "VocabularyFilterName": "string", "VocabularyName": "string" } }, "LanguageOptions": [ "string" ], "Media": { "MediaFileUri": "string", "RedactedMediaFileUri": "string" }, "MediaFormat": "string", "MediaSampleRateHertz": number, "ModelSettings": { "LanguageModelName": "string" }, "Settings": { "ChannelIdentification": boolean, "MaxAlternatives": number, "MaxSpeakerLabels": number, "ShowAlternatives": boolean, "ShowSpeakerLabels": boolean, "VocabularyFilterMethod": "string", "VocabularyFilterName": "string", "VocabularyName": "string" }, "StartTime": number, "Subtitles": { "Formats": [ "string" ], "OutputStartIndex": number, "SubtitleFileUris": [ "string" ] }, "Tags": [ { "Key": "string", "Value": "string" } ], "ToxicityDetection": [ { "ToxicityCategories": [ "string" ] } ], "Transcript": { "RedactedTranscriptFileUri": "string", "TranscriptFileUri": "string" }, "TranscriptionJobName": "string", "TranscriptionJobStatus": "string" } }

Response Elements

If the action is successful, the service sends back an HTTP 200 response.

The following data is returned in JSON format by the service.

TranscriptionJob

Provides detailed information about the current transcription job, including job status and, if applicable, failure reason.

Type: TranscriptionJob object

Errors

For information about the errors that are common to all actions, see Common Errors.

BadRequestException

Your request didn't pass one or more validation tests. This can occur when the entity you're trying to delete doesn't exist or if it's in a non-terminal state (such as IN PROGRESS). See the exception message field for more information.

HTTP Status Code: 400

ConflictException

A resource already exists with this name. Resource names must be unique within an AWS account.

HTTP Status Code: 400

InternalFailureException

There was an internal error. Check the error message, correct the issue, and try your request again.

HTTP Status Code: 500

LimitExceededException

You've either sent too many requests or your input file is too long. Wait before retrying your request, or use a smaller file and try your request again.

HTTP Status Code: 400

See Also

For more information about using this API in one of the language-specific AWS SDKs, see the following: