Is your feature request related to a problem? Please describe.
The main issue I've faced with the current TTS integration is later reqs failing for long responses due to how content's split up by sentence. Long enough responses are split into enough reqs that they exceed OAI's ratelimits & fail.
Describe the solution you'd like
Since OAI's voice models are limited by RPM (e.g., tts-1-hd's limited to 5 RPM @ tier 2), it'd be great to see some more advanced options surfaced in the Admin Panel UI for controlling how responses are split up / sent out to the selected TTS API.
Originally created by @kiosion on GitHub (Aug 22, 2024).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/4834
**Is your feature request related to a problem? Please describe.**
The main issue I've faced with the current TTS integration is later reqs failing for long responses due to how content's split up by sentence. Long enough responses are split into enough reqs that they exceed OAI's ratelimits & fail.
**Describe the solution you'd like**
Since OAI's voice models are limited by RPM (e.g., `tts-1-hd`'s limited to 5 RPM @ tier 2), it'd be great to see some more advanced options surfaced in the Admin Panel UI for controlling how responses are split up / sent out to the selected TTS API.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @kiosion on GitHub (Aug 22, 2024).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/4834
Is your feature request related to a problem? Please describe.
The main issue I've faced with the current TTS integration is later reqs failing for long responses due to how content's split up by sentence. Long enough responses are split into enough reqs that they exceed OAI's ratelimits & fail.
Describe the solution you'd like
Since OAI's voice models are limited by RPM (e.g.,
tts-1-hd's limited to 5 RPM @ tier 2), it'd be great to see some more advanced options surfaced in the Admin Panel UI for controlling how responses are split up / sent out to the selected TTS API.