Forum Discussion
Slow down entire slide/voice - text to speech
Is there an easy way to slow down the text-to-speech for each slide-to-speech file?
Meaning, I don't want to slow down a specific word or words but rather the entire paragraph of text on a slide. I want to slow down the entire rate for that voice file.
Any way to do this?
It would be nice to have this an option when the "insert text to speech" window opens. Then have a selection to slow down the rate at that point, with options of rate speeds)
8 Replies
Hello Katrina,
Thanks for reaching out! I'm happy to share that you can slow down your text-to-speech audio using SSML, which Storyline supports.
To enable SSML support, you must enclose your text-to-speech narration with the start tag <speak> and the end tag </speak>. You can control the speaking rate using the <prosody> tag, as shown by the sample below and the attached file containing the resulting text-to-speech audio. If you'd like to slow down an entire paragraph, enclose all of it with the <prosody> tag.
<speak>
<prosody rate="x-slow">
We’re so glad you’re here! You are now part of a growing community of professionals creating, collaborating, and connecting worldwide via Articulate Storyline.Whether you’ve joined to create something of your own or just to connect with other professionals, we’ve got something for you.
</prosody>
</speak>Please don't hesitate to reach out if you have additional questions.
I have found that these tags don't work consistently. The first time I used them the speech slowed down. Subsequent updates to pronunication, did not slow down. How can I make them do so?
Hello FelicityWigh485! In this case, formatting does matter. Do you have extra spaces, line breaks, or trailing whitespace? That can prevent tags from being recognized. If you can copy what you've included in your text-to-speech, we can test it on our end and provide steps from there!
Hello,
I see you mentioned Storyline, but I’m wondering if there’s an easy way to slow down the text-to-speech when using Rise's AI audio. I’ve noticed that when composing the text, I try to format it like a teleprompter script for novice broadcasters by using ellipses and commas, as well as the Return button, to approximate how I’d like it to sound.
However, I'm seeking advice on overall pacing. I’ve tried adjusting the sliders for Stability, Similarity, and Style Exaggeration, but I still find myself generating about 10 to 15 attempts before I get the audio to sound just right without needing to edit it further. Do you have any tips or suggestions?
Thank you!
Hello AmEy_Michelle,
Thanks for reaching out! Rise 360, AI-generated text-to-speech does not fully support SSML (Speech Synthesis Markup Language) like Storyline 360 does. You can still influence the pacing by using punctuation like commas and ellipses, as you've been doing.
If you need to control pacing manually, you can add pauses using the syntax `<break time="1.5s" />` in your script, but keep in mind that excessive use of break tags can cause instability.
Another workaround is to use Storyline 360 to generate your audio with SSML tags and then export the audio file for use in your Rise 360 course. Here's a look at this in action for your reference.
I'll open this discussion up to our community to share examples of what has worked well in their projects!
If the tags aren't consistently supported and the software doesn't allow you to get there easily (15 tries is scream-at-the-sky-level frustration), you know what to do: bust that file into Tenacity, slow it down to whatever speed you want, and then reimport it to your SL file. Rinse and repeat for each speed you want.
