[GH-ISSUE #8525] feat: Add Speed input for Admin Panel -> Settings -> Audio: TTS Settings #15157

New Issue

GiteaMirror · 2026-04-19T21:27:03-05:00

GiteaMirror commented

2026-04-19 21:27:03 -05:00

Originally created by @SamuraiBarbi on GitHub (Jan 13, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/8525

Feature Request

Admin Panel -> Settings -> Audio

TTS Settings
When setting the Text-to-Speech Engine to OpenAI, we're able to specify the host, api key, TTS Voice, and TTS Model, however we have no way to specify Speed aka how fast the text is spoken.
Can we please add a Speed input so that requests sent to an OpenAI compatible TTS server include a speed parameter?

Here's an example of speed of 1.25 being included in such a sequest. When I send a speed arg/value to my local TTS server the audio I get back is spoken faster.

curl -X POST http://localhost:8085/v1/audio/speech -H "Content-Type: application/json" -d '{"model": "kokoro", "voice": "bf_isabella", "speed": 1.25, "input": "A total of 931 inmates are currently working around the clock, cutting fire lines and removing fuel from behind structures to slow the spread of California’s devastating wildfires, according to the California Department of Corrections and Rehabilitation", "output_format": "mp3"}' -o output.mp3

Originally created by @SamuraiBarbi on GitHub (Jan 13, 2025). Original GitHub issue: https://github.com/open-webui/open-webui/issues/8525 # Feature Request Admin Panel -> Settings -> Audio TTS Settings When setting the Text-to-Speech Engine to OpenAI, we're able to specify the host, api key, TTS Voice, and TTS Model, however we have no way to specify Speed aka how fast the text is spoken. Can we please add a Speed input so that requests sent to an OpenAI compatible TTS server include a speed parameter? Here's an example of speed of 1.25 being included in such a sequest. When I send a speed arg/value to my local TTS server the audio I get back is spoken faster. ```bash curl -X POST http://localhost:8085/v1/audio/speech -H "Content-Type: application/json" -d '{"model": "kokoro", "voice": "bf_isabella", "speed": 1.25, "input": "A total of 931 inmates are currently working around the clock, cutting fire lines and removing fuel from behind structures to slow the spread of California’s devastating wildfires, according to the California Department of Corrections and Rehabilitation", "output_format": "mp3"}' -o output.mp3 ```

GiteaMirror closed this issue

2026-04-19 21:27:03 -05:00

GiteaMirror referenced this issue

2026-04-19 23:15:44 -05:00

[GH-ISSUE #15157] Issue: Web search has no effect on LLM output #17487

GiteaMirror referenced this issue

2026-04-25 06:52:56 -05:00

[GH-ISSUE #15157] Issue: Web search has no effect on LLM output #33016

GiteaMirror referenced this issue

2026-05-05 18:50:11 -05:00

[GH-ISSUE #15157] Issue: Web search has no effect on LLM output #56153

Sign in to join this conversation.