Forum Discussion
Storyline text-to-speech AI Audio Voices - SSML Tags
As per title, are SSML Tags compatible with the new text-to-speech AI voices?
I was trying to change the phonetic pronunciation of a specific word, to no avail. I was however successful in adding a pause with a tag.
Thanks!
The <break time="#s"/> function does not seem to work for more than 1 second (voice Matilda). I tried 2s, 3s, 4s, and 6s. Additionally, it does not work if you repeat the command. It seems the only way to slow down the TTS is to break it up into smaller pieces and move them on the timeline. This is a huge limitation, not having SSML support.
Hi @ReganKleiner,
Thanks for reaching out and I'm sorry to hear you've hit this snag!
We're tracking a possible software bug where AI Text-to-speech speeds up the audio if there are too many SSML break tags. I've linked this discussion so we can update you when we have news to share. In the meantime, if you're comfortable sharing your file, we'd be happy to take a closer look for further testing. You can upload it here or privately in a support case.
I'd like an answer too. I tried using <break time="250ms"/>, but it didn't work. How did you add a pause?
I used this code:
<speak>
This is a test to know if <break time="3s"/>pauses work correctly with the voice.</speak>
You should hear the pause if you try it. Always make sure to start and end your text with <speak> and </speak>, respectively.
Thank you!
I want to use a break time that's less than one second. <break time="250ms"/> and <break time=".7s"/> doesn't seem to work.
Hello MEJohnson,
I am happy to chime in. I understand you'd like to use an SSML "break" code with milliseconds.
<break time="250ms"/> should be the correct format, but 250 milliseconds may be too short to be noticeable. If you want to add a 0.7-second break, please try the following SSML code:
Please don't hesitate to reach out if you have additional questions. If you want more targeted help, you can also connect with us privately in a support case.
Thank you!
You're welcome, MEJohnson!
I'd like to know an answer to the original question as well. Without the ability to change pronunciation, I can't see myself using this over other options.
Hello WrennCorcoran,
Happy to help!
SSML tags aren't recognized by the AI assistant when generating AI audio. AI Assistant has limited support for speech synthesis markup language (SSML) because AI-generated voices are designed to understand the relationship between words and adjust delivery accordingly. I'll add you to an open feature request we have to include SSML tags in the list of supported features of the AI assistance and we'll let you know when it makes it to our product roadmap!Great! I would also appreciate this functionality. The new AI voices are amazing, but adding SSML compatibility to them would rock!
Try the substute tag
here is an example:
<speak>
<break time="500ms"/>
Best practices voltage testing include.
<sub alias="lyve 2 dead 2 lyve">Live-dead-live</sub> <break time="125ms"/>testing makes sure the meter is working correctly before and after a test.<break time="250ms"/>
</speak>
