[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/runtime.py

@@ -33,7 +33,7 @@ from .protocols import AudioTransport, LlmProvider, SttProvider, TtsProvider, Wa
 from .stt import CloudAsrSttProvider, SherpaOnnxSttProvider, is_valid_transcript_text
 from .transport import SoundDeviceAudioTransport
 from .tts import CloudTtsProvider, MacSayTtsProvider, SentenceBuffer
-from .vad import EnergyVadProvider, HybridVadProvider, SherpaOnnxVadProvider, VadRecorder
+from .vad import EnergyVadProvider, HybridVadProvider, PrimarySpeakerVadRecorder, SherpaOnnxVadProvider, VadRecorder
 from .wakeword import SherpaOnnxKeywordWakeWordProvider
 
 
@@ -345,6 +345,23 @@ def build_live_runtime(config: AppConfig, reporter: RuntimeReporter | None = Non
         vad_provider = SherpaOnnxVadProvider(config.speech_models_dir, threshold=config.vad_threshold)
     else:
         vad_provider = EnergyVadProvider()
+    recorder_cls = PrimarySpeakerVadRecorder if config.endpoint_mode == "primary_speaker" else VadRecorder
+    recorder_kwargs = {
+        "provider": vad_provider,
+        "min_duration_ms": config.vad_min_duration_ms,
+        "end_silence_ms": config.vad_end_silence_ms,
+        "no_speech_timeout_ms": config.vad_no_speech_timeout_ms,
+        "max_recording_ms": config.vad_max_recording_ms,
+    }
+    if recorder_cls is PrimarySpeakerVadRecorder:
+        recorder_kwargs.update(
+            {
+                "speaker_profile_ms": config.speaker_profile_ms,
+                "speaker_absent_ms": config.speaker_absent_ms,
+                "similarity_threshold": config.speaker_similarity_threshold,
+                "min_rms": config.speaker_min_rms,
+            }
+        )
     return VoiceAssistantPipeline(
         config=config,
         transport=SoundDeviceAudioTransport(output_device=config.audio_output_device),
@@ -355,13 +372,7 @@ def build_live_runtime(config: AppConfig, reporter: RuntimeReporter | None = Non
             threshold=config.wake_kws_threshold,
             score=config.wake_kws_score,
         ),
-        vad_recorder=VadRecorder(
-            vad_provider,
-            min_duration_ms=config.vad_min_duration_ms,
-            end_silence_ms=config.vad_end_silence_ms,
-            no_speech_timeout_ms=config.vad_no_speech_timeout_ms,
-            max_recording_ms=config.vad_max_recording_ms,
-        ),
+        vad_recorder=recorder_cls(**recorder_kwargs),
         stt=stt,
         llm=OpenAICompatibleLlmProvider(config),
         tts=tts,