I have been searching for a decent and low-cost lifelike text-to-speech service or software for commercial use. After checking some recommendations, they are either not available internationally, difficult to install via command line, or simply too expensive e.g. more than ten or twenty per month. Then I came across Amazon Polly in a Quora answer.
Criteria for speech-to-text service or software:
- Free or affordable
- Decent speech output
- Commercial use allowed
- Not difficult to install or use
- Mac and preferably cross-plaforms
How affordable is Amazon Polly?
Amazon Polly is affordable, especially for the average user. It only costs $4 per million characters for Standard voices and $16 per million characters for Neural voices. And you get started with millions of characters for free per month for 12 months in the free tier. Neural voices sound more subtly natural than Standard ones, but both are acceptable for common usage.
For example, a YouTube animation video that runs for about 3-5 minutes that narrates up to 2500 characters costs only 4 cents for neural voice. The whole novel of A Christmas Carol by Charles Dickens of almost 4 hours of narration for only $2.64.
Be sure to check if you are within the limits of the free tier here and use it moderately for a month and wait for the charges to come in the next month.
For example, I used Amazon Polly to create about 3 short slides videos, and the service charge is about USD0.13.
Amazon Polly for commercial use?
Yes, it can be used for commercial videos (including YouTube).
How good is Amazon Polly text-to-speech output?
Both Standard (8 voices) and Neural (9 voices) output are good or excellent.
For example, listen to these short pieces of Standard and Neural voices for comparisons.
I would recommend Neural voices for most.
Is Amazon Polly easy to use?
Amazon Polly is a breeze to use.
There are some self-explanatory options but basically, just type or paste the text into the input text box, and click Download.
Bookmarking the Amazon Polly Dashboard URL for future quick access can be a time-saving routine.
Is there an example of a YouTube video with Amazon Polly narration?
Can Amazon Polly be used on Mac and Windows?
Yes, in fact not just Mac and Windows but any computers including Linux with an internet connection since it is accessible from a browser.
What may Amazon Polly be used for?
Any legitimate business and personal use that requires human-like text-to-speech service. Some ideas may include website services that provide support for sight-disabled users, reading services for educational purposes, blog to YouTube video creation, and others.
Amazon Polly is truly a gem among the text-to-speech services and software I checked out, in fact, it is the only one that met all my criteria. If your business, blog, or YouTube videos may benefit from a text-to-speech narration service, then consider Amazon Polly for a legitimate and cost-effective solution.
I would recommend Neural over Standard voice for any good business and personal usage.
Are there any recommended speech apps?