LLMUserContextAggregator: reduce aggregation_timeout to 0.5

CHANGELOG.md

@@ -25,6 +25,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Changed
 
+- Changed the user aggregator timeout for late transcriptions from 1.0s to 0.5s
+  (`LLMUserAggregatorParams.aggregation_timeout`). Sometimes, the STT services
+  might give us more than one transcription which could come after the user
+  stopped speaking. We still want to include these additional transcriptions
+  with the first one because it's part of the user turn. This is what this
+  timeout is helpful with.
+
 - Short utterances not detected by VAD while the bot is speaking are now
   ignored. This reduces the amount of bot interruptions significantly providing
   a more natural conversation experience.

src/pipecat/processors/aggregators/llm_response.py

@@ -50,7 +50,7 @@ from pipecat.utils.time import time_now_iso8601
 
 @dataclass
 class LLMUserAggregatorParams:
-    aggregation_timeout: float = 1.0
+    aggregation_timeout: float = 0.5
 
 
 @dataclass