From 164574c1460f2d7ef8407d9c954263317c25234b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rapha=C3=ABl?=
 <203084086+raphael-bynome@users.noreply.github.com>
Date: Sun, 8 Nov 2026 10:47:19 +0100
Subject: [PATCH] docs: fix WHISPER_COMPUTE_TYPE and BYPASS_PYDUB_PREPROCESSING
 defaults

---
 .../audio/speech-to-text/env-variables.md                   | 2 +-
 docs/reference/env-configuration.mdx                        | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/docs/features/chat-conversations/audio/speech-to-text/env-variables.md b/docs/features/chat-conversations/audio/speech-to-text/env-variables.md
index 63d4fcf602..479d71ce85 100644
--- a/docs/features/chat-conversations/audio/speech-to-text/env-variables.md
+++ b/docs/features/chat-conversations/audio/speech-to-text/env-variables.md
@@ -23,7 +23,7 @@ Most of these settings can also be configured in the **Settings → Admin → Au
 
 | Variable | Description | Default |
 |----------|-------------|---------|
-| `BYPASS_PYDUB_PREPROCESSING` | Skip pydub-based preprocessing (MP3 conversion, compression, chunk splitting) before sending audio to the STT engine. Applies to all engines. Useful when the upstream provider already handles these steps, or when ffmpeg is unavailable on the host. | `false` |
+| `BYPASS_PYDUB_PREPROCESSING` | Skip pydub-based preprocessing (MP3 conversion, compression, chunk splitting) before sending audio to the STT engine. Applies to all engines. Useful when the upstream provider already handles these steps, or when ffmpeg is unavailable on the host. | `false` (always `true` on the slim image) |
 
 ### Local Whisper
 
diff --git a/docs/reference/env-configuration.mdx b/docs/reference/env-configuration.mdx
index f97b333d41..96ffd3dbc6 100644
--- a/docs/reference/env-configuration.mdx
+++ b/docs/reference/env-configuration.mdx
@@ -5833,7 +5833,7 @@ Note: If none of the specified languages are available and `en` was not in your
 #### `BYPASS_PYDUB_PREPROCESSING`
 
 - Type: `bool`
-- Default: `False`
+- Default: `False` (always `True` on the slim image)
 - Description: When enabled, skips pydub-based preprocessing (format conversion to MP3, compression, and chunked splitting by size) before audio is sent to the Speech-to-Text engine. Useful when the upstream provider already handles these steps, when ffmpeg is unavailable on the host, or when you want to pass audio through untouched. Applies to all STT engines.
 
 ### Whisper Speech-to-Text (Local)
@@ -5854,8 +5854,8 @@ Note: If none of the specified languages are available and `en` was not in your
 #### `WHISPER_COMPUTE_TYPE`
 
 - Type: `str`
-- Default: `int8` (CPU), `float16` (CUDA)
-- Description: Sets the compute type for Whisper model inference. Defaults to `int8` for CPU and `float16` for CUDA (with fallback to `int8/int8_float16`).
+- Default: `int8`
+- Description: Sets the compute type for Whisper model inference. The default is `int8` on every device, including the `:cuda` image: Open WebUI does not switch it automatically. On a GPU, set it explicitly (for example `float16`).
 
 #### `WHISPER_VAD_FILTER`