In this video, I review the top 5 best audio narration options for your eLearning courses (#1 and #2 might surprise you).
Links to these options:
Amazon Polly: https://aws.amazon.com/polly/
Kelly Labrecque: https://www.kellylabrecquevoiceover.com/
Microsoft TTS: http://bit.ly/MStts
Neospeech/Readspeaker: https://www.readspeaker.com/
WellSaid: https://www.wellsaidlabs.com
I personally am a fan of such innovative technology. WellSaid labs are my favorite, they sound really natural to me. My team and I (elai.io) integrated with them and I like how the videos are narrated.
Judging by the quality of these text to speech systems, they’re still not there yet compared to any actual narration. They’re getting there but they’re still a way off. I find them all more than a little off putting.
The decision to use text to speech is usually an economical one. For example, a freelance elearning developer may not have a quality microphone or the proper space in which to record their narration. Text to speech, especially if it’s included in your authoring tool has few barriers for that elearning developer.
One problem I have experienced with these is they seem to ramp up English language accents – seemingly because they want to offer “US” or “UK” versions you then get unrealistic or annoying voices. Would be much better to offer something with a minimal accent, especially when selling to an international audience. What have people found works best for a toned down accent?
A voice-over artist is best but when you can get away with ai voices like https://wellsaidlabs.com you can save time and money on not having to buy studio time for a professional recording. Also, corrections to your script can be easily implemented.
Such a helpful video. Thanks for posting!
You’re welcome.
I love Polly! I recently started to use Wellsaid Labs…heard of them Paul? They are the closest I’ve gotten to an AI sounding “human.”
I also go to Fiverr to hire voice actors if looking for something specific. I can try to put up a sample this week of a Wellsaid Labs sample, but they likely have some on their site if you’d like to listen. It is paid though 🙁
I guess you haven’t watch the video.
No, should’ve stated I didn’t watch, I just saw the mention of Amazon Polly…and had I reviewed the resources I would’ve saw you’re familiar! Lol
LOL
Hi Paul,
Yes, I agree with you. Nothing beats live, professional narrators. For the last decade I have been using narrators for my “paid” elearning projects. I generally use two narrators (male and female) presenting information and discussing back and forth. Feedback from learners has been very positive, and clients are more than willing to pay the extra fees. The challenge is writing the narration scripts for two, making it sound natural and engaging.
In the last couple of years, I have also used Amazon Polly fairly extensively. I prefer the Neural voices and occasionally use SSML for emphasis, pauses, loudness, etc. I have found that using two voices (male and female), either alternating on paragraphs or slides, keeps the level of attention and interest up. Amazon hasn’t asked for any payment (yet).
In your video, regarding Amazon Polly, you state “There doesn’t seem to be a cost associated with it …”
According to the Amazon Polly site:
“With Amazon Polly, you only pay for what you use. You are charged based on the number of characters of text that you convert either to speech or to Speech Marks metadata. In addition, you can cache and replay Amazon Polly’s generated speech at no additional cost. It’s easy to get started with the Amazon Polly Free Tier, try it today.”
Thanks for the additional details.
You must be logged in to post a comment.
