From d883edc832b501bd539021d0b4df5be163b4efbb Mon Sep 17 00:00:00 2001
From: "github-actions[bot]" <github-actions[bot]@users.noreply.github.com>
Date: Wed, 6 May 2026 14:43:14 +0000
Subject: [PATCH] docs: update Deepgram Flux STT for watchdog_min_timeout
 parameter

Add watchdog_min_timeout parameter to both DeepgramFluxSTTService and
DeepgramFluxSageMakerSTTService documentation, and update the watchdog
note to reflect the new dynamic threshold behavior.
---
 api-reference/server/services/stt/deepgram.mdx | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/api-reference/server/services/stt/deepgram.mdx b/api-reference/server/services/stt/deepgram.mdx
index f0a0fb73..b921c0bf 100644
--- a/api-reference/server/services/stt/deepgram.mdx
+++ b/api-reference/server/services/stt/deepgram.mdx
@@ -290,6 +290,12 @@ Supports the standard [service connection events](/api-reference/server/events/s
   Whether the bot should be interrupted when Flux detects user speech.
 </ParamField>
 
+<ParamField path="watchdog_min_timeout" type="float" default="0.5">
+  Minimum silence duration in seconds before the watchdog sends silence to
+  prevent dangling turns. The actual threshold is `max(chunk_duration * 2,
+  watchdog_min_timeout)`, adapting automatically to the audio chunk size in use.
+</ParamField>
+
 ### Settings
 
 Runtime-configurable settings passed via the `settings` constructor argument using `DeepgramFluxSTTService.Settings(...)`. These can be updated mid-conversation with `STTUpdateSettingsFrame`. See [Service Settings](/pipecat/fundamentals/service-settings) for details.
@@ -545,6 +551,12 @@ Supports the standard [service connection events](/api-reference/server/events/s
   Whether to interrupt the bot when Flux detects user speech.
 </ParamField>
 
+<ParamField path="watchdog_min_timeout" type="float" default="0.5">
+  Minimum silence duration in seconds before the watchdog sends silence to
+  prevent dangling turns. The actual threshold is `max(chunk_duration * 2,
+  watchdog_min_timeout)`, adapting automatically to the audio chunk size in use.
+</ParamField>
+
 <ParamField
   path="settings"
   type="DeepgramFluxSageMakerSTTService.Settings"
@@ -634,7 +646,7 @@ await task.queue_frame(
 - **EagerEndOfTurn**: Enabling `eager_eot_threshold` provides faster response times by predicting end-of-turn before it is confirmed. EagerEndOfTurn transcripts are pushed as `InterimTranscriptionFrame`s. If the user resumes speaking, a `TurnResumed` event is fired.
 - **Multilingual support**: Use the `flux-general-multi` model with `language_hints` to bias transcription toward specific languages (EN, ES, FR, DE, HI, RU, PT, JA, IT, NL). `TranscriptionFrame.language` reflects the detected language for each turn. Omit hints for auto-detection or pass a subset to bias toward expected languages.
 - **SageMaker deployment**: Requires a Deepgram Flux model deployed to an AWS SageMaker endpoint. Unlike Nova models, Flux provides native turn detection and does not require external VAD.
-- **No KeepAlive needed**: The Flux protocol uses a watchdog mechanism that sends silence when needed to maintain the connection, so manual KeepAlive messages are not required.
+- **No KeepAlive needed**: The Flux protocol uses a dynamic watchdog mechanism that sends silence when needed to maintain the connection (threshold: `max(chunk_duration * 2, watchdog_min_timeout)`), so manual KeepAlive messages are not required.
 
 ### Event Handlers