[GH-ISSUE #5557] [BUG]: Piper TTS issue #5163

Closed
opened 2026-06-05 14:52:23 -04:00 by yindo · 1 comment
Owner

Originally created by @trancemuter on GitHub (Apr 29, 2026).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5557

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

is there any way to make the bult in Piper tts not read out loud the markdown special characters? it is so annoying to listen piper saying "Asterix" all the time, it defeats the purpose of the whole voice system, Also, is there a way to stream the TTS output instead waiting until the whole response loads and then glitch out if it is too long? these are pretty basic issues that I am very surprised it is not solved after so many updates... I am praying with every update that this is solved... doesn't anybody use the voice feature at AnythingLLM? Doesn't this annoy the hell out the crew? why isn't this addressed? other web UIs have solved this long ago even on mobile... what is the solution to this?

Are there known steps to reproduce?

No response

LLM Provider & Model (if applicable)

Ollama

Embedder Provider & Model (if applicable)

No response

Originally created by @trancemuter on GitHub (Apr 29, 2026). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5557 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? is there any way to make the bult in Piper tts not read out loud the markdown special characters? it is so annoying to listen piper saying "Asterix" all the time, it defeats the purpose of the whole voice system, Also, is there a way to stream the TTS output instead waiting until the whole response loads and then glitch out if it is too long? these are pretty basic issues that I am very surprised it is not solved after so many updates... I am praying with every update that this is solved... doesn't anybody use the voice feature at AnythingLLM? Doesn't this annoy the hell out the crew? why isn't this addressed? other web UIs have solved this long ago even on mobile... what is the solution to this? ### Are there known steps to reproduce? _No response_ ### LLM Provider & Model (if applicable) Ollama ### Embedder Provider & Model (if applicable) _No response_
yindo added the possible bug label 2026-06-05 14:52:23 -04:00
yindo closed this issue 2026-06-05 14:52:23 -04:00
Author
Owner

@timothycarambat commented on GitHub (Apr 29, 2026):

Also, is there a way to stream the TTS output instead waiting until the whole response loads and then glitch out if it is too long?

Yeah this is annoying. It is unfortunately non-trivial, but we are going to be fixing this as that whole part of the desktop code needs a rewrite which is why it sucks so badly. I do not use the STT honestly because I can read faster than it can speak - its not my main use case.

I know we have some vision-limited users, but they use their own screen readers instead since it needs to work across the whole UI - which is likely why this has stay so low priority. Will fix the markdown thing (among some other conditions) to prevent this going forward.

<!-- gh-comment-id:4348171777 --> @timothycarambat commented on GitHub (Apr 29, 2026): > Also, is there a way to stream the TTS output instead waiting until the whole response loads and then glitch out if it is too long? Yeah this is annoying. It is unfortunately non-trivial, but we are going to be fixing this as that whole part of the desktop code needs a rewrite which is why it sucks so badly. I do not use the STT honestly because I can read faster than it can speak - its not my main use case. I know we have some vision-limited users, but they use their own screen readers instead since it needs to work across the whole UI - which is likely why this has stay so low priority. Will fix the markdown thing (among some other conditions) to prevent this going forward.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#5163