Jump to Content

New API Documentation - Developer Preview Available

We are excited to announce the developer preview of our new API documentation for AWS SDK for JavaScript v3. Please follow instructions on the landing page to leave us your feedback.

A word, phrase, or punctuation mark in your transcription output, along with various associated attributes, such as confidence score, type, and start and end times.


  • MedicalItem


Confidence?: number

The confidence score associated with a word or phrase in your transcript.

Confidence scores are values between 0 and 1. A larger value indicates a higher probability that the identified item correctly matches the item spoken in your media.

Content?: string

The word or punctuation that was transcribed.

EndTime?: number

The end time, in milliseconds, of the transcribed item.

Speaker?: string

If speaker partitioning is enabled, Speaker labels the speaker of the specified item.

StartTime?: number

The start time, in milliseconds, of the transcribed item.

Type?: string

The type of item identified. Options are: PRONUNCIATION (spoken words) and PUNCTUATION.