Amazon Polly
AWS 文档中描述的 AWS 服务或功能可能因区域而异。要查看适用于中国区域的差异,请参阅中国的 AWS 服务入门


Synthesizes UTF-8 input, plain text or SSML, to a stream of bytes. SSML input must be valid, well-formed SSML. Some alphabets might not be available with all the voices (for example, Cyrillic might not be read at all by English voices) unless phoneme mapping is used. For more information, see How it Works.

Request Syntax

POST /v1/speech HTTP/1.1 Content-type: application/json { "Engine": "string", "LanguageCode": "string", "LexiconNames": [ "string" ], "OutputFormat": "string", "SampleRate": "string", "SpeechMarkTypes": [ "string" ], "Text": "string", "TextType": "string", "VoiceId": "string" }

URI Request Parameters

The request does not use any URI parameters.

Request Body

The request accepts the following data in JSON format.


Specifies the engine (standard or neural) for Amazon Polly to use when processing input text for speech synthesis. Using a voice that is not supported for the engine selected will result in an error.

Type: String

Valid Values: standard | neural

Required: No


Optional language code for the Synthesize Speech request. This is only necessary if using a bilingual voice, such as Aditi, which can be used for either Indian English (en-IN) or Hindi (hi-IN).

If a bilingual voice is used and no language code is specified, Amazon Polly will use the default language of the bilingual voice. The default language for any voice is the one returned by the DescribeVoices operation for the LanguageCode parameter. For example, if no language code is specified, Aditi will use Indian English rather than Hindi.

Type: String

Valid Values: arb | cmn-CN | cy-GB | da-DK | de-DE | en-AU | en-GB | en-GB-WLS | en-IN | en-US | es-ES | es-MX | es-US | fr-CA | fr-FR | is-IS | it-IT | ja-JP | hi-IN | ko-KR | nb-NO | nl-NL | pl-PL | pt-BR | pt-PT | ro-RO | ru-RU | sv-SE | tr-TR

Required: No


List of one or more pronunciation lexicon names you want the service to apply during synthesis. Lexicons are applied only if the language of the lexicon is the same as the language of the voice. For information about storing lexicons, see PutLexicon.

Type: Array of strings

Array Members: Maximum number of 5 items.

Pattern: [0-9A-Za-z]{1,20}

Required: No


The format in which the returned output will be encoded. For audio stream, this will be mp3, ogg_vorbis, or pcm. For speech marks, this will be json.

When pcm is used, the content returned is audio/pcm in a signed 16-bit, 1 channel (mono), little-endian format.

Type: String

Valid Values: json | mp3 | ogg_vorbis | pcm

Required: Yes


The audio frequency specified in Hz.

The valid values for mp3 and ogg_vorbis are "8000", "16000", "22050", and "24000". The default value for standard voices is "22050". The default value for neural voices is "24000".

Valid values for pcm are "8000" and "16000" The default value is "16000".

Type: String

Required: No


The type of speech marks returned for the input text.

Type: Array of strings

Array Members: Maximum number of 4 items.

Valid Values: sentence | ssml | viseme | word

Required: No


Input text to synthesize. If you specify ssml as the TextType, follow the SSML format for the input text.

Type: String

Required: Yes


Specifies whether the input text is plain text or SSML. The default value is plain text. For more information, see Using SSML.

Type: String

Valid Values: ssml | text

Required: No


Voice ID to use for the synthesis. You can get a list of available voice IDs by calling the DescribeVoices operation.

Type: String

Valid Values: Aditi | Amy | Astrid | Bianca | Brian | Camila | Carla | Carmen | Celine | Chantal | Conchita | Cristiano | Dora | Emma | Enrique | Ewa | Filiz | Geraint | Giorgio | Gwyneth | Hans | Ines | Ivy | Jacek | Jan | Joanna | Joey | Justin | Karl | Kendra | Kimberly | Lea | Liv | Lotte | Lucia | Lupe | Mads | Maja | Marlene | Mathieu | Matthew | Maxim | Mia | Miguel | Mizuki | Naja | Nicole | Penelope | Raveena | Ricardo | Ruben | Russell | Salli | Seoyeon | Takumi | Tatyana | Vicki | Vitoria | Zeina | Zhiyu

Required: Yes

Response Syntax

HTTP/1.1 200 Content-Type: ContentType x-amzn-RequestCharacters: RequestCharacters AudioStream

Response Elements

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

The response returns the following HTTP headers.


Specifies the type audio stream. This should reflect the OutputFormat parameter in your request.

  • If you request mp3 as the OutputFormat, the ContentType returned is audio/mpeg.

  • If you request ogg_vorbis as the OutputFormat, the ContentType returned is audio/ogg.

  • If you request pcm as the OutputFormat, the ContentType returned is audio/pcm in a signed 16-bit, 1 channel (mono), little-endian format.

  • If you request json as the OutputFormat, the ContentType returned is audio/json.


Number of characters synthesized.

The response returns the following as the HTTP body.


Stream containing the synthesized speech.



This engine is not compatible with the voice that you have designated. Choose a new voice that is compatible with the engine or change the engine and restart the operation.

HTTP Status Code: 400


The specified sample rate is not valid.

HTTP Status Code: 400


The SSML you provided is invalid. Verify the SSML syntax, spelling of tags and values, and then try again.

HTTP Status Code: 400


The language specified is not currently supported by Amazon Polly in this capacity.

HTTP Status Code: 400


Amazon Polly can't find the specified lexicon. This could be caused by a lexicon that is missing, its name is misspelled or specifying a lexicon that is in a different region.

Verify that the lexicon exists, is in the region (see ListLexicons) and that you spelled its name is spelled correctly. Then try again.

HTTP Status Code: 404


Speech marks are not supported for the OutputFormat selected. Speech marks are only available for content in json format.

HTTP Status Code: 400


An unknown condition has caused a service failure.

HTTP Status Code: 500


SSML speech marks are not supported for plain text-type input.

HTTP Status Code: 400


The value of the "Text" parameter is longer than the accepted limits. For the SynthesizeSpeech API, the limit for input text is a maximum of 6000 characters total, of which no more than 3000 can be billed characters. For the StartSpeechSynthesisTask API, the maximum is 200,000 characters, of which no more than 100,000 can be billed characters. SSML tags are not counted as billed characters.

HTTP Status Code: 400

See Also

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