Top 10 features of Amazon Transcribe

In this post, I will tell you about the top 10 features of Amazon Transcribe that you must know to fully utilize the power of this Machine Learning service. I hope you will find them useful while working with Amazon Transcribe.


Hello Everyone, In this post, I am going to tell you about the top 10 features of Amazon Transcribe that you must know to fully utilize the power of this Machine Learning service. I hope you are excited to know about them, So, Without wasting any time, Let’s get started!

Real-time Transcription

With this feature you can generate transcription for your speech in real-time, you just need to click on Start Streaming button and start speaking to your microphone, Amazon Transcribe will show you the transcribed text live on screen.

This feature is useful for generating subtitle information for a live video or an online meeting.

Transcription Job

This is one of the prime features of Amazon Transcribe for doing batch processing of multiple media files simultaneously. You can define a Transcription Job by selecting multiple parameters to get the desired text output. This option helps in automatically transcribing hundreds of audio files at the same time.

Useful in auto transcribing a video course, a series of podcasts, or any type of audio content in bulk form with minimum efforts.

Automatic Language Detection

With this feature Amazon Transcribe can automatically detect the language of your input media file. So that you don’t have to specify the language for each file manually.

This feature is useful when you want to auto transcribe multiple media files of various languages all together.

Channel Identification

If your audio input file has Voices on multiple channels then you can use Channel Identification feature of Amazon Transcribe. With this feature, Amazon Transcribe will automatically identify the voices on each channel and will generate the transcripts for them. The transcription output will be having a combined text for the whole of the recording and also the separate text corresponding to each channel.

Very useful in the scenario where customer and support agent voices are recorded on separate channels.

Speaker Identification

This feature helps in identifying multiple speakers in an audio file, So if you want to auto transcribe a media file that is having multiple speakers talking in it then you can use this speaker identification feature of Amazon Transcribe. The output will be having the whole text and the separate texts corresponding to each speaker in the media file. You can choose to identify up to 10 speakers in an audio file.

This feature can be used to auto transcribe a conference call where multiple participants were involved.

Custom Vocabulary

With this feature, you can provide a list of specific words that you want Amazon Transcribe to recognize in your audio input. These are generally domain-specific words and phrases, words that Amazon Transcribe isn’t recognizing, or proper nouns. You can have up to 100 vocabularies in your account, the size limit for a Custom Vocabulary is 50Kb. You can have it defined in either a list format or a table format.

This feature can be used to help Amazon Transcribe understand and place industry-specific terms correctly in the transcription output, like one of these AWS service names EC2, S3, RDS etc.

Vocabulary Filtering

This feature helps in masking or completely removing profane words automatically in the transcription output. You can choose to mask in-appropriate words with three asterisks *** or remove them completely, by defining the vocabulary filter for unwanted words or terms.

Very useful in masking or removing offensive terms while generating subtitles for a family based video or show.

Automatic Content Redaction

This feature automatically redacts sensitive personally identifiable information also known as PII, like the name of a person, his bank account number, credit card info, etc. from the transcription results and replaces those parts with the text PII.

It is very useful for meeting Security and Compliance requirements. Because by using this feature you can protect your Customer’s sensitive information.

Alternative Results

With this feature, we can request Amazon Transcribe to give us the alternate transcription results even with somewhat low confidence scores while transcribing a media file, by default it only gives us the transcription results that have the highest confidence level to deliver the high accuracy. We can request maximum 10 alternative results in the transcription output.

It is useful while doing the manual review of transcription results to choose the most appropriate transcription results.

Job Queuing

This feature is helpful in submitting a large number of transcription jobs to Amazon Transcribe, there is a limit of running 100 transcription jobs simultaneously and if you want to run more than this number then you can submit the transcription jobs using Job Queuing, this will put additional transcription jobs in the queue and will process them as soon as the slot becomes available, by default there are 100 slots for transcription jobs, which means you can submit up to 10,000 jobs to the queue.

This feature is useful in submitting a large number of transcription jobs to auto transcribe 1000s of files.


So, That was all about the Top 10 features of Amazon Transcribe that makes this service very powerful. If you are interested in learning them further then you can explore one of my courses at

Thank you for going through this post and giving me an opportunity to share this valuable information with you.

Let me know in the comments which feature out of these Top 10 features of Amazon Transcribe you liked the most?

What is Amazon Transcribe and How to use it?

Want to learn What is Amazon Transcribe, How you can use it, What is the pricing related to it and What are some of it’s Use Cases, then read this quick post to get answers to all of your queries.


Hello Everyone, In this post, I am going to tell you about What is Amazon Transcribe, How we can use it in a simple and easy way possible, What is the pricing related to it, What are the possible use cases where we can use it, Finally I will conclude this article by giving you some thoughts on how to take Amazon Transcribe output to the next level and some suggestions for further reading on this topic. I hope you are excited, So, Let’s get started!

What is Amazon Transcribe?

Amazon Transcribe is an Automatic Speech Recognition(ASR) Service by Amazon available on the AWS Cloud platform. It is capable of recognizing speech from an existing audio or video file. It can also recognize speech from the streaming audio.

You simply need to provide the media file in one of the supported formats and the service will deliver the transcribed text automatically.

It is well know as Speech-to-text service or STT for short.

How we can use Amazon Transcribe?

The simplest way to use Amazon Transcribe is from AWS Management Console.

Let’s explore how to use Real-time transcription using Amazon Transcribe and for this,

  1. You simply need to go to Amazon Transcribe
  2. Choose Real-time transcription
  3. Choose desired language
  4. Click on Start streaming, this will ask to access your Microphone, allow that and start speaking

5. When you start speaking you will be able to see the transcribed text getting displayed in real-time under transcription output area.

6. Once you have done speaking, you will be able to download the transcribed text by clicking on Download full transcript button.

Let’s now see how to create a Transcription Job to transcribe a media file that is uploaded to an S3 storage bucket.

  1. Go to Amazon Transcribe
  2. Click on Transcription Jobs on left hand side
  3. Now, Click on Create job
  4. Give a name to the Transcription job
  5. Keep rest of the settings to default

6. Select desired language of the media file

7. Now, Under the Input data, browse the media file that must be already uploaded into an S3 storage bucket, earlier we have to manually give the path of media file existing under an S3 storage bucket, but now with recent updates we can browse the contents of an S3 storage bucket and select the desired media file for generating the transcription

8. Under Output data keep default option of “Service-managed S3 bucket” this option will help us to see the transcribed output on the screen itself, otherwise we have to manually go to the S3 storage bucket to fetch the output data

9. Click on Next

10. On this next screen, most of the settings are optional, we can skip them for the time being and directly click on Create button to create the Transcription job

11. The Transcription job will go into the status of In progress and after a short while it will go into Complete status and once the Transcription job is Complete, we can Click on Download full transcript button to see the output on the screen itself

Amazon Transcribe Pricing

Amazon Transcribe follows a pay-as-you-go model for pricing based on the duration of audio transcribed in a month.

Under Free Tier, You can transcribe 60 minutes of audio without any cost every month for first 12 months.

Beyond Free Tier, You have to pay $0.0004 per second, with a minimum per request charge of 15 seconds.

Pricing for Amazon Transcribe may vary based upon the region that you select, So for accurate pricing for your AWS region please refer to this AWS pricing link related to it.

Amazon Transcribe Possible Use Cases

Amazon Transcribe can be used in any of the use cases where you need to convert audio, video, or speech data into text form,

for example:

  1. In contact center – To covert customer and agent call recordings into text
  2. In meetings and conferences – To automatically generate the Minutes of Meetings using Real-time transcription feature

3. In video production – To generate the subtitles for a recorded or live video

4. In advertising – To show relevant ads to users based upon the video content they have consumed

5. In search and compliance – To convert audio and video data into text to make the content easily searchable and make it compliant with desired regulations

Taking Output of Amazon Transcribe to Next Level

So, you have transcribed your audio or video files to text, Now you can take that text to the next level, You can translate the text from one language to another using Amazon Translate, and after translating you can give voice to that text using Amazon Polly in the native accents of your end-users. That way you can produce content in one language but can deliver it to global audiences in multiple languages, that too with their native accents to expand the reach of your products.


Amazon Transcribe is very easy and affordable to use, anyone with some basic computer knowledge can learn to use it.

If you are interested in learning all of the great features that Amazon Transcribe offers than you can check out one of my courses related to it available at, I will show you how you can utilize Amazon Transcribe in a step-by-step manner from AWS Management Console, Command Line Interface and also using Python SDK.

Further Reading

  1. If you want to know how to create your AWS Free Tier account then you can watch all the steps in this Video
  2. If you want to see a quick video demonstration of Amazon Transcribe then you can watch this Video

Lastly, Thank you for going through this post and giving me an opportunity to share this information with you.

Do let me know for which use case you would like to use Amazon Transcribe?