[主说话人端点]：完成音色消失结束录音，包含临时音色画像、端点配置和回归测试

src/owner_voice_pet/config.py

@@ -26,6 +26,12 @@ class AppConfig:
     wake_kws_score: float = 1.0
     wake_ack_text: str = "我在"
     post_playback_drain_ms: int = 50
+    pipeline_mode: str = "live_turn_based"
+    endpoint_mode: str = "primary_speaker"
+    speaker_profile_ms: int = 600
+    speaker_absent_ms: int = 300
+    speaker_similarity_threshold: float = 0.70
+    speaker_min_rms: float = 0.012
     vad_provider: str = "hybrid"
     vad_threshold: float = 0.5
     vad_min_duration_ms: int = 250
@@ -37,6 +43,7 @@ class AppConfig:
     tts_model: str = "mimo-v2.5-tts"
     tts_voice: str = "mimo_default"
     speech_models_dir: Path = Path("models")
+    context_mode: str = "session_memory"
     context_max_messages: int = 12
     context_max_chars: int = 12000
 
@@ -67,6 +74,14 @@ class AppConfig:
             wake_kws_score=float(get("WAKE_KWS_SCORE", "1.0") or "1.0"),
             wake_ack_text=get("WAKE_ACK_TEXT", "我在") or "我在",
             post_playback_drain_ms=int(get("POST_PLAYBACK_DRAIN_MS", "50") or "50"),
+            pipeline_mode=(get("PIPELINE_MODE", "live_turn_based") or "live_turn_based").lower(),
+            endpoint_mode=(get("ENDPOINT_MODE", "primary_speaker") or "primary_speaker").lower(),
+            speaker_profile_ms=int(get("SPEAKER_PROFILE_MS", "600") or "600"),
+            speaker_absent_ms=int(get("SPEAKER_ABSENT_MS", "300") or "300"),
+            speaker_similarity_threshold=float(
+                get("SPEAKER_SIMILARITY_THRESHOLD", "0.70") or "0.70"
+            ),
+            speaker_min_rms=float(get("SPEAKER_MIN_RMS", "0.012") or "0.012"),
             vad_provider=(get("VAD_PROVIDER", "hybrid") or "hybrid").lower(),
             vad_threshold=float(get("VAD_THRESHOLD", "0.5") or "0.5"),
             vad_min_duration_ms=int(get("VAD_MIN_DURATION_MS", "250") or "250"),
@@ -78,6 +93,7 @@ class AppConfig:
             tts_model=get("TTS_MODEL", "mimo-v2.5-tts") or "mimo-v2.5-tts",
             tts_voice=get("TTS_VOICE", "mimo_default") or "mimo_default",
             speech_models_dir=Path(get("SPEECH_MODELS_DIR", "models") or "models"),
+            context_mode=(get("CONTEXT_MODE", "session_memory") or "session_memory").lower(),
             context_max_messages=int(get("CONTEXT_MAX_MESSAGES", "12") or "12"),
             context_max_chars=int(get("CONTEXT_MAX_CHARS", "12000") or "12000"),
         )
@@ -178,6 +194,60 @@ class AppConfig:
                     "startup",
                 )
             )
+        if self.pipeline_mode not in {"live_turn_based"}:
+            errors.append(
+                ProviderError(
+                    ErrorCode.CONFIG_MISSING_VALUE,
+                    "OWNER_PIPELINE_MODE must be live_turn_based",
+                    False,
+                    "config",
+                    "startup",
+                )
+            )
+        if self.endpoint_mode not in {"primary_speaker", "vad"}:
+            errors.append(
+                ProviderError(
+                    ErrorCode.CONFIG_MISSING_VALUE,
+                    "OWNER_ENDPOINT_MODE must be primary_speaker or vad",
+                    False,
+                    "config",
+                    "startup",
+                )
+            )
+        for name, value in {
+            "OWNER_SPEAKER_PROFILE_MS": self.speaker_profile_ms,
+            "OWNER_SPEAKER_ABSENT_MS": self.speaker_absent_ms,
+        }.items():
+            if value <= 0:
+                errors.append(
+                    ProviderError(
+                        ErrorCode.CONFIG_MISSING_VALUE,
+                        f"{name} must be positive",
+                        False,
+                        "config",
+                        "startup",
+                    )
+                )
+        if not 0 < self.speaker_similarity_threshold <= 1:
+            errors.append(
+                ProviderError(
+                    ErrorCode.CONFIG_MISSING_VALUE,
+                    "OWNER_SPEAKER_SIMILARITY_THRESHOLD must be in (0, 1]",
+                    False,
+                    "config",
+                    "startup",
+                )
+            )
+        if self.speaker_min_rms <= 0:
+            errors.append(
+                ProviderError(
+                    ErrorCode.CONFIG_MISSING_VALUE,
+                    "OWNER_SPEAKER_MIN_RMS must be positive",
+                    False,
+                    "config",
+                    "startup",
+                )
+            )
         if self.vad_provider not in {"hybrid", "local", "energy"}:
             errors.append(
                 ProviderError(
@@ -224,6 +294,16 @@ class AppConfig:
                     "startup",
                 )
             )
+        if self.context_mode not in {"session_memory"}:
+            errors.append(
+                ProviderError(
+                    ErrorCode.CONFIG_MISSING_VALUE,
+                    "OWNER_CONTEXT_MODE must be session_memory",
+                    False,
+                    "config",
+                    "startup",
+                )
+            )
         return errors
 
     def api_url(self, path: str) -> str: