In this article, I am going to show you how to use Amazon Transcribe (automatic speech recognition service), to create a text transcript of a pre-recorded speech file in English, after uploading it to a S3 bucket using the AWS Management Console.

Amazon Transcribe

• It is a very easy and a useful tool for creating transcriptions of audio data, either a media file uploaded in an Amazon S3 bucket or a media stream, and converts it to text data.

• It can transcribe public speeches, business meeting notes, customer calls, broadcast TV, on-demand videos, class lectures and perform medical transcription in real-time.

• You can use Amazon Transcribe as a standalone service or to add speech-to-text capabilities to any application.

• You can transcribe from these languages list

• Transcription jobs are of 2 types:

• Batch transcription jobs - Media files stored in an Amazon S3 bucket

• Streaming transcription jobs - Media streams in real time.

Please visit my GitHub Repository for S3 articles on various topics being updated on constant basis.

Lets get started!

Objectives:

1. Create a S3 bucket

2. Upload an audio file into S3 bucket

3. Create transcription job

4. Review transcription results

Pre-requisites:

• AWS user account with admin access, not a root account.
• Create an IAM role, with AmazonS3FullAccess

Resources Used:

Amazon Transcribe

IAM Access Policy

S3 Bucket

Steps for implementation to this project:

1. Create a S3 bucket

On Amazon S3 console / Create bucket / Under General configuration /

Bucket name: - oprah-audio

AWS Region: - US East (N. Virginia) us-east-1

• Take all defaults and Create bucket

2. Upload an audio file into S3 bucket

• Click on your buckets name to navigate to the bucket / On the Buckets Home page / Select Upload / Add files / Upload the oprah-audio.mp3 file

Upload

• Select oprah-audio.mp3 file / Under Properties / For Object overview / Copy the S3 URL / Save it for future use

s3://oprah-audio/oprah-audio.mp3

3. Create transcription job

• From the top menu bar, select Services then begin typing Transcribe in the search bar and select Amazon Transcribe to open the service console.

• On the Amazon Transcribe Console / Transcription jobs page, click Create job / Under Specify job details / Job settings /

Name: - oprah-audio-transcribe-job

Language: - English,US (en-US)

Input data / Input file location on S3: - s3://oprah-audio/oprah-audio.mp3

Output data location type: take the default - Service-managed S3 bucket.

Subtitle type format: -

• Amazon Transcribe supports WebVTT (VTT) and SubRip (SRT) file types.

• In theSubtitle file formatfield, you can choose either or both file types for output.

• If you select both types, you get two files that are exported to the same S3 bucket.

• I am not not using either formats.

Next

• On the Configure page / Under Customization / Custom vocabulary

This feature helps you to recognize words and phrases that are specific to your application. I am not choosing this feature as I am not using any application.

• Create job

4. Review transcription results

• After the Transcription jobs - shows Complete / Click on oprah-audio-transcribe-job / Under Transcription preview / Text

• you can see the following transcribed text

• If the transcribed text is long, you have to to scroll down to the Transcription panel to view the whole transcription job output.

Cleanup

• Delete the audio file - oprah-audio.mp3

• Delete the S3 bucket

• Delete the Transcription job

What we have done so far

We have used Amazon Transcribe to create a text transcript of a pre-recorded speech file in English, after uploading it to a S3 bucket using the AWS Management Console.