[持续对话判定]：完成恢复待机延迟修复，包含本地规则收敛和分类器跳过回归测试

src/owner_voice_pet/continuation.py

@@ -35,28 +35,27 @@ class ContinuationDecisionProvider(Protocol):
 
 class RuleContinuationDecisionProvider:
     _continue_patterns = (
-        "你想",
-        "你要",
-        "你需要",
-        "需要我",
-        "要不要",
-        "是否需要",
         "可以告诉我",
-        "告诉我",
         "请告诉我",
-        "你希望",
-        "你更想",
-        "哪一个",
-        "哪一部分",
-        "哪种",
-        "哪个",
         "请选择",
         "选一个",
         "请补充",
-        "需要补充",
         "补充一下",
         "继续吗",
     )
+    _ambiguous_continue_patterns = (
+        "我还可以",
+        "我可以继续",
+        "还可以继续",
+        "可以继续",
+        "继续展开",
+        "继续讲",
+        "继续聊",
+        "两个方向",
+        "三个方向",
+        "几个方向",
+        "几个部分",
+    )
     _standby_patterns = (
         "这是",
         "已经",
@@ -70,6 +69,11 @@ class RuleContinuationDecisionProvider:
         "抱歉",
         "出错",
         "失败",
+        "先这样",
+        "到这里",
+        "随时叫我",
+        "有需要再叫我",
+        "需要时再叫我",
     )
 
     def decide(
@@ -88,7 +92,9 @@ class RuleContinuationDecisionProvider:
             return ContinuationDecision("standby", 0.82, "assistant appears to finish the answer", "rule")
         if len(text) <= 18 and not text.endswith(("?", "？", "吗", "呢")):
             return ContinuationDecision("standby", 0.72, "short non-question reply", "rule")
-        return ContinuationDecision("unknown", 0.0, "rule uncertain", "rule")
+        if any(pattern in text for pattern in self._ambiguous_continue_patterns):
+            return ContinuationDecision("unknown", 0.0, "ambiguous continuation offer", "rule")
+        return ContinuationDecision("standby", 0.78, "assistant did not ask for immediate input", "rule")
 
 
 class LlmContinuationDecisionProvider:

tests/test_continuation.py

@@ -58,6 +58,24 @@ class ContinuationDecisionTests(unittest.TestCase):
         self.assertEqual(decision.action, "standby")
         self.assertEqual(len(llm.calls), 1)
 
+    def test_hybrid_generic_completed_reply_does_not_call_llm_classifier(self) -> None:
+        llm = FakeClassifierLlm('{"action":"continue","confidence":0.99,"reason":"不应调用"}')
+        provider = HybridContinuationDecisionProvider(
+            RuleContinuationDecisionProvider(),
+            LlmContinuationDecisionProvider(llm, threshold=0.65),
+            threshold=0.65,
+        )
+
+        decision = provider.decide(
+            user_text="没有呢",
+            assistant_text="明白了，我先保持待机。有需要再叫我就行。",
+            history=[],
+        )
+
+        self.assertEqual(decision.action, "standby")
+        self.assertEqual(decision.provider, "rule")
+        self.assertEqual(llm.calls, [])
+
     def test_hybrid_high_confidence_llm_can_continue(self) -> None:
         llm = FakeClassifierLlm('{"action":"continue","confidence":0.91,"reason":"等待用户选择"}')
         provider = HybridContinuationDecisionProvider(

tests/test_live_runtime.py

@@ -594,6 +594,38 @@ class LiveRuntimeTests(unittest.TestCase):
         self.assertEqual(event_types[-1], STANDBY_RESUMED)
         self.assertEqual(len(llm.calls), 1)
 
+    def test_completed_reply_returns_to_standby_without_cloud_classifier_delay(self) -> None:
+        frames = [wake_frame(0, 0)]
+        frames.extend(segment_frames(1, 20, partials=["没有呢", "没有呢"]))
+        transport = MemoryAudioTransport(frames, flush_clears_input=False)
+        llm = QueueLlmProvider([["明白了，我先保持待机。有需要再叫我就行。"]])
+        runtime = VoiceAssistantPipeline(
+            config=AppConfig(
+                llm_api_key="secret",
+                speech_provider="cloud",
+                wake_ack_text="",
+                followup_listen_timeout_ms=3000,
+            ),
+            transport=transport,
+            wakeword=KeywordWakeWordProvider(),
+            vad_recorder=VadRecorder(EnergyVadProvider(), min_duration_ms=40, end_silence_ms=40),
+            stt=QueueSttProvider(["没有呢"]),
+            realtime_stt=MetadataSttProvider(),
+            llm=llm,
+            tts=SineTtsProvider(),
+            context=ConversationContext(),
+            reporter=RecordingReporter(),
+            event_bus=PipelineEventBus(),
+        )
+
+        summary = runtime.run(max_turns=1)
+        event_types = [event.type for event in runtime.event_bus.events]
+
+        self.assertEqual(summary.completed_turns, 1)
+        self.assertNotIn(FOLLOWUP_LISTENING, event_types)
+        self.assertEqual(event_types[-1], STANDBY_RESUMED)
+        self.assertEqual(len(llm.calls), 1)
+
     def test_barge_in_interrupts_playback_and_keeps_only_spoken_context(self) -> None:
         first_question_frames = [wake_frame(0, 0)]
         first_question_frames.extend(segment_frames(1, 20, partials=["第一问", "第一问"]))