Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-19 14:13:35 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #2240] [BUG]: PiperTTS issues #1464

New Issue
Closed
opened 2026-02-22 18:24:56 -05:00 by yindo · 6 comments
yindo commented 2026-02-22 18:24:56 -05:00
Owner
Copy Link
Copy Source

Originally created by @ghost on GitHub (Sep 7, 2024).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/2240

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

Firstly, sorry for being frustrated last time I gave a bug report. I love your software.

Okay, so PiperTTS likes to work with short text, and that is fine. Interestingly, If the text is longer I will get
AnythingLLM v1 6 4 PiperTTS - Screenshot 2024-09-07 172056
However, as soon as I select the next large text it plays the first text that errored out. Furthermore, it likes to add the word "asterisks" between each paragraph and any keyboard character besides "the period". I've flush voice cache a ton and doesn't seem to help.

If there is no fix, it's fine don't worry about it. You have come a long way and this is spectacular software!

Windows 10
Llama 3.1 Storm q8 (all features work with this model, sorry about prior complaints)
AnythingLLM v1.6.4

Are there known steps to reproduce?

Using PiperTTS and normal sized output from Llama 3 which is usually two paragraphs or 350 words. After it errors out, select another.

Originally created by @ghost on GitHub (Sep 7, 2024). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/2240 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? Firstly, sorry for being frustrated last time I gave a bug report. I love your software. Okay, so PiperTTS likes to work with short text, and that is fine. Interestingly, If the text is longer I will get ![AnythingLLM v1 6 4 PiperTTS - Screenshot 2024-09-07 172056](https://github.com/user-attachments/assets/59711d00-f73c-44ce-a7a9-e5ced05c2243) However, as soon as I select the next large text it plays the first text that errored out. Furthermore, it likes to add the word "asterisks" between each paragraph and any keyboard character besides "the period". I've flush voice cache a ton and doesn't seem to help. If there is no fix, it's fine don't worry about it. You have come a long way and this is spectacular software! Windows 10 Llama 3.1 Storm q8 (all features work with this model, sorry about prior complaints) AnythingLLM v1.6.4 ### Are there known steps to reproduce? Using PiperTTS and normal sized output from Llama 3 which is usually two paragraphs or 350 words. After it errors out, select another.
yindo added the possible bugcore-team-onlyDesktopinvestigating labels 2026-02-22 18:24:56 -05:00
yindo closed this issue 2026-02-22 18:24:56 -05:00
yindo commented 2026-02-22 18:24:56 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Sep 9, 2024):

Will see if we can replicate it. If you can share system specs on your end that would help a lot - the worker system should have timeout disabled by default so longer-processes can run, like longer text predictions.

Clearly it does eventually load since you get the result after, just not on the first click.

@timothycarambat commented on GitHub (Sep 9, 2024): Will see if we can replicate it. If you can share system specs on your end that would help a lot - the worker system should have timeout disabled by default so longer-processes can run, like longer text predictions. Clearly it does eventually load since you get the result after, just not on the first click.
yindo commented 2026-02-22 18:24:57 -05:00
Author
Owner
Copy Link
Copy Source

@bvqbao commented on GitHub (Sep 14, 2024):

I have this error too. Version 1.6.6 on Windows 11.
In my case, I can't even play a sample voice.
image

@bvqbao commented on GitHub (Sep 14, 2024): I have this error too. Version 1.6.6 on Windows 11. In my case, I can't even play a sample voice. <img width="534" alt="image" src="https://github.com/user-attachments/assets/488242d8-88f6-46c9-8b16-198a4ff94013">
yindo commented 2026-02-22 18:24:57 -05:00
Author
Owner
Copy Link
Copy Source

@TwoPintOhh commented on GitHub (Nov 18, 2024):

I have this error too. Version 1.6.6 on Windows 11. In my case, I can't even play a sample voice. image

Im having the same issue. i think its because of my low internet speed it needs a little longer to download the model. once im lucky and get the tick, im all good for that model. Im using docker so i cant figure out how to import the model manualy

@TwoPintOhh commented on GitHub (Nov 18, 2024): > I have this error too. Version 1.6.6 on Windows 11. In my case, I can't even play a sample voice. <img alt="image" width="534" src="https://private-user-images.githubusercontent.com/4357219/367529345-488242d8-88f6-46c9-8b16-198a4ff94013.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.Lpgc8vQ-I7EY1ufll2lLi0uL49EJwqb42skEy_RU968"> Im having the same issue. i think its because of my low internet speed it needs a little longer to download the model. once im lucky and get the tick, im all good for that model. Im using docker so i cant figure out how to import the model manualy
yindo commented 2026-02-22 18:24:57 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Feb 18, 2025):

This issue is due to streaming of assets from HF timing out within the 30s worker limit. This can be because:

  • User access to HF is blocked (geo blocked by HF)
  • Internet speeds too slow
@timothycarambat commented on GitHub (Feb 18, 2025): This issue is due to streaming of assets from HF timing out within the 30s worker limit. This can be because: - User access to HF is blocked (geo blocked by HF) - Internet speeds too slow
yindo commented 2026-02-22 18:24:57 -05:00
Author
Owner
Copy Link
Copy Source

@yabo-boye commented on GitHub (Feb 24, 2025):

any way of making it possible to download the asset in a bigger limit? i am having this problem also and i think its befcuase of my 10mb/s internet connection

@yabo-boye commented on GitHub (Feb 24, 2025): any way of making it possible to download the asset in a bigger limit? i am having this problem also and i think its befcuase of my 10mb/s internet connection
yindo commented 2026-02-22 18:24:57 -05:00
Author
Owner
Copy Link
Copy Source

@Bino5150 commented on GitHub (Jan 25, 2026):

This issue is due to streaming of assets from HF timing out within the 30s worker limit. This can be because:

* User access to HF is blocked (geo blocked by HF)

* Internet speeds too slow

I'm having the same issues with the Piper timeout error on anything longer than a quick sentence or two from the LLM. I'm running AnythingLLM v1.10.0 with LM Studio. I'm running completely local. I can play the voice in the preview setting, and it works just fine for shorter responses.

HP Zbook Studio G7 on Linux Mint (i7, 16GB, Nvidia Quadro T1000 4GB vram)

The LLM and the Piper model are stored and running locally.

@Bino5150 commented on GitHub (Jan 25, 2026): > This issue is due to streaming of assets from HF timing out within the 30s worker limit. This can be because: > > * User access to HF is blocked (geo blocked by HF) > > * Internet speeds too slow I'm having the same issues with the Piper timeout error on anything longer than a quick sentence or two from the LLM. I'm running AnythingLLM v1.10.0 with LM Studio. I'm running completely local. I can play the voice in the preview setting, and it works just fine for shorter responses. HP Zbook Studio G7 on Linux Mint (i7, 16GB, Nvidia Quadro T1000 4GB vram) The LLM and the Piper model are stored and running locally.
yindo changed title from [BUG]: PiperTTS issues to [GH-ISSUE #2240] [BUG]: PiperTTS issues 2026-06-05 14:40:53 -04:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Desktop
Docker
Integration Request
Integration Request
OS: Linux
OS: Mobile
OS: Windows
UI/UX
blocked
bug
bug
core-team-only
documentation
duplicate
embed-widget
enhancement
feature request
github_actions
good first issue
investigating
needs info / can't replicate
possible bug
pull-request
Mirrored from GitHub Pull Request
 question
stage: specifications
wontfix
No Label Desktop core-team-only investigating possible bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#1464
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?