Skip to main content
Participant
May 15, 2026

Using Audio - Generate Speech: Credits Consumed, No Speech File Generated

  • May 15, 2026
  • 1 reply
  • 15 views
  1. Pasted text into window.
  2. Selected voice.
  3. Added emotion effects prompts to the text window (ctrl+A to select all, used ‘amused’).
  4. Slowed the voice speed to 95%.
  5. Drop the pitch one stop above bottom.
  6. Selected the first four lines of text to play and test settings.
  7. No additional adjustments required.
  8. Selected ‘Generate’ for 38 credits.

The generating speech status window popped up. Then ended abruptly after maybe 10 seconds. It usually takes up to a minute or more so I knew something was amiss. But I waited. It has now been about 15 minutes and still no file generated.

This is a ‘Community’ Forum, so I suppose it is unlikely my credits will be refunded via this bug report. A visible process for that might be enhancement request.

    1 reply

    May 16, 2026

    Hey ​@BeMoreBetter 

    Could you attempt using a smaller amount of text? Consider breaking up the text into 3 bite sized pieces. This will make it easier to move and edit when you make your video as well.

    Let us know if that helps :)

    Cheers

    Nate

    Participant
    May 20, 2026

    Great Suggestion! It also allows me to zero in the emotional inflections.

    Thanks for offering this solution!

     

    With no intention to diminish my gratitude in any way, I must that the back and forth, cut and paste is only a work-around, and not viable for long-form presentations. My particular use-case is generating the voice for a colleague with disabilities who delivers presentations and trainings on the ‘dis-life’ (living with disabilities). The pieces we have developed range from 3-5 minutes to over an hour.

    The 5 minute piece I am currently working on is now spread across 6 files. Counting all edits and regenerations, about 10 files, I think. This is about the maximum project length appropriate given the limitations. Keeping track of… your tracks, copy end points through changes in breakpoints, etc. What I appreciated about it was how easy it made modulating tone and emotion. But keeping the overall structure organized was only manageable given the small size. Any larger and it would quickly have gotten out of hand.

    I understand that the feature I am working with here is beta, so I hope my comments are received in the spirit delivered. Given my experience, I think the ideal workspace for this would be something closer to how Adobe has implemented subtitles/captions in the Premiere.

    I will leave that feedback as a feature request next, but essentially the user would input the full working text which the system would convert it to text blocks, allowing the user the granular control experienced in this work-around but minus the difficulties of cut & paste, generated file confusion, etc.

     

    But again, N8, those are only my opinions on the functionality and/or perceived limitations of the current state and my opinion on a possible future solution. You provided me with what I needed today- a way to finish my current project.

    BIG Thank You!

    And don’t sell yourself short, my friend. In my book, you’re a N10!

    Peace.

    BeMo