mirror of
https://github.com/open-webui/open-webui.git
synced 2026-05-22 09:31:58 -05:00
enh: Add TTS Response Splitting option for splitting on markdown (bold/italics) #3479
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: github-starred/open-webui#3479
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @TheRealPSV on GitHub (Jan 29, 2025).
Feature Request
Is your feature request related to a problem? Please describe.
The Response Splitting option for TTS is helpful, but the "Sentences" option doesn't split off bold/italicized sentences that some models will output, creating run-on sentences when sent to the TTS engine.
Describe the solution you'd like
Create a Response Splitting option similar to the Sentences option, but that also splits on bold or italics markers (asterisks) for models that insert "roleplaying" sentences, such as characters' actions, to help prevent run-on sentences in TTS.
Describe alternatives you've considered
Another potential option could be to have the markdown cleanup that is performed during sending to TTS split the sentence as it removes markdown, but a user-configurable option would likely be more robust, as certain models are more likely to require this, while it may be detrimental to others.
Additional context
Example of a paragraph that becomes a run-on sentence when passed to TTS:
Hello! My name is Henry. Holds out hand Nice to meet you.
In the above paragraph, the Sentence option's split will result in the TTS engine outputting the run-on sentence "Holds out hand nice to meet you."