Amazon AI

en flag
fr flag
de flag
it flag
ja flag
es flag
sv flag
Voiced by Amazon Polly

While exploring the features within AWS, I came across Amazon AI, which can perform text-to-speech conversion. While there is nothing really new about it, it was something interesting to try out. I ended up doing a bulk update on all my previous posts, and as you’ll notice you can now have the AI read to you the article instead. You also have the option to subscribe to the audio feed.

What is Amazon AI?

Amazon Polly is a service that turns text into lifelike speech, allowing you to create applications that talk, and build entirely new categories of speech-enabled products. Amazon Polly is a Text-to-Speech service that uses advanced deep learning technologies to synthesize speech that sounds like a human voice.

Amazon’s explanation of the Polly service
Amazon Polly demo video

Thoughts on Amazon AI

I have read comments and reviews saying they are not all that impressed by the voice sound, but I still think it’s interesting, and it was fun to set up. The pricing itself doesn’t look bad either and based on the amount of text I’ll be doing each month I doubt it will cost much of anything for me each month. For the time being, I will include it with each of my posts, as well as explore the option of translations with it.

Amazon, for the first year, will allow you 5,000,000 characters at no cost per month. Even after that first year, the cost to convert any of my posts would be pennies at most. I think in the end it may be more of a novelty, and after the year I may stop using it. I do like the translation service, as it will translate not only the voice portion but the text as well. Therefore I will enable this feature as well and see how it works. I’m not fluent enough in any other language however to tell how accurate the translations are though. So if you have thoughts on the translation quality, be sure and let me know.

Alexa Skill

The option that I think gives the AI the most usefulness is by adding an Alexa Skill that interfaces with the website. The skill pulls the audio feeds from the S3 bucket in AWS and Alexa will read the blog posts to you. I have already created the skill, however, at this time it is pointed to a test site. I just need to edit it to point to this site and then anyone with an Amazon echo device can listen to the blog just by saying, “Alexa, open the root user“. If you wish to use this skill, you can download it from the Amazon Alexa Store.


Ivan Windon

Ivan Windon is a Lead Engineer at Verizon Enterprise Solutions. Ivan is actively engaged in Cloud Technologies with AWS, Google, and Azure. Ivan has extensive experience with Linux and Windows administration, DNS, Networking, IDM, and Security. In his free time, he enjoys being with his wife and two children. The family enjoys hiking, and traveling when able. His favorite locations are Yosemite NPS, and San Francisco, California.

You may also like...

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: