Start Synthesis Task and Poll to Completion
A 2-step recipe with Amazon Polly. We can help you get access.
Recipe · 2 steps
Start Synthesis Task and Poll to Completion
with Amazon Polly
Start an asynchronous synthesis task and poll it until it reaches a terminal state.
- Start an asynchronous speech synthesis task that writes its output audio to the supplied S3 bucket, returning a task id to poll on.
- Poll the synthesis task by id, branching back to poll again while the task is still scheduled or in progress.
What you'll need to have ready
- amzDate: The X-Amz-Date timestamp used to sign the requests.
- authorization: The full SigV4 Authorization header value for the request.
- contentSha256: The X-Amz-Content-Sha256 hex digest of the request payload.
- securityToken: Optional X-Amz-Security-Token for temporary credentials.
- engine: Engine to use for synthesis (standard or neural).
- text: The input text (plain text or SSML) to synthesize.
- textType: Whether the input text is plain text or ssml.
- outputFormat: The audio output format (mp3, ogg_vorbis, pcm, or json).
- outputS3BucketName: The S3 bucket name to which the output file will be saved.
- outputS3KeyPrefix: Optional S3 key prefix for the output speech file.
- voiceId: Voice ID to use for the synthesis (e.g. Joanna).
For your developer
Arazzo workflow amazon-polly:start-synthesis-task-poll, 2 operations: StartSpeechSynthesisTask → GetSpeechSynthesisTask. Specs on APIs.io.