Media Services: Use AWS AI services to automate captioning & subtitling

In this lab, you prototype a workflow which takes a video with English audio, automatically generates the a transcript from the audio using Amazon Transcribe, translates the English text into subtitles for another language using Amazon Translate, and uses those subtitle text files to generate an alternative audio track using Amazon Polly. You will use AWS Elemental MediaConvert to combine the video, audio tracks, captions, and subtitles into a single streamable distribution that allows end users to select their desired audio track and which if any captions/subtitles to display.

Level

Intermediate

Duration

Course Objectives

In this course, you will learn how to:

• Construct a process to automate the generation of captions, alternate language subtitles, and alternate language audio tracks using three AWS AI services – Amazon Transcribe, Amazon Translate, and Amazon Polly
• Generate a transcript of a video using Amazon Transcribe
• Translate the transcript into an alternate language using Amazon Translate
• Transform the alternate language transcript into alternate language audio using Amazon Polly
• Construct an Apple HLS stack using AWS Elemental MediaConvert
• Validate that audio and captions/subtitles work as intended in a video player

Intended Audience

This course is intended for:

• Architects

Prerequisites

We recommend that attendees of this course have the following prerequisites:

• Familiar with basic navigation of the AWS Management Console 
• Comfortable editing scripts using a text editor

Course Outline

• Task 1: Generate a transcript of a video using Amazon Transcribe
• Task 2: Translate transcript into an alternate language using Amazon Translate
• Task 3: Transform the alternate language transcript into alternate language audio using Amazon Polly
• Task 4: Construct an Apple HLS stack using AWS Elemental MediaConvert
• Task 5: Validate that audio and captions/subtitles work as intended in a video player