请问下realtime api的VAD功能可以用吗？ #20
Description
options=RtcOptions(
channel_name=self.call_sid,
uid=666,
sample_rate=16000,
channels=1,
enable_pcm_dump=False,
enable_vad=True,
vad_configs=vad_config
),
我这样传会报这个错
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ AttributeError: 'dict' object has no attribute 'pre_start_recognize_count' Exception ignored on calling ctypes callback function: <bound method AudioFrameObserverInner._on_playback_audio_frame_before_mixing of <agora.rtc._ctypes_handle._audio_frame_observer.AudioFrameObserverInner object at 0x16c678bd0>> Traceback (most recent call last): File "/Users/lavrock/work/ai_call/.venv/lib/python3.12/site-packages/agora/rtc/_ctypes_handle/_audio_frame_observer.py", line 102, in _on_playback_audio_frame_before_mixing vad_result_state, vad_result_bytes = self._vad_instance_manager.process(channel_id_str, user_id_str, frame) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/lavrock/work/ai_call/.venv/lib/python3.12/site-packages/agora/rtc/audio_vad_manager.py", line 49, in process self._add_vad_instance(channel_id, user_id) File "/Users/lavrock/work/ai_call/.venv/lib/python3.12/site-packages/agora/rtc/audio_vad_manager.py", line 30, in _add_vad_instance self._instance_map[key] = AudioVadV2(self._vad_config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/lavrock/work/ai_call/.venv/lib/python3.12/site-packages/agora/rtc/voice_detection.py", line 60, in __init__ self._start_size = self._vad_configure.pre_start_recognize_count + self._vad_configure.start_recognize_count ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ AttributeError: 'dict' object has no attribute 'pre_start_recognize_count' Exception ignored on calling ctypes callback function: <bound method AudioFrameObserverInner._on_playback_audio_frame_before_mixing of <agora.rtc._ctypes_handle._audio_frame_observer.AudioFrameObserverInner object at 0x16c678bd0>> Traceback (most recent call last): File "/Users/lavrock/work/ai_call/.venv/lib/python3.12/site-packages/agora/rtc/_ctypes_handle/_audio_frame_observer.py", line 102, in _on_playback_audio_frame_before_mixing vad_result_state, vad_result_bytes = self._vad_instance_manager.process(channel_id_str, user_id_str, frame) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/lavrock/work/ai_call/.venv/lib/python3.12/site-packages/agora/rtc/audio_vad_manager.py", line 49, in process self._add_vad_instance(channel_id, user_id) File "/Users/lavrock/work/ai_call/.venv/lib/python3.12/site-packages/agora/rtc/audio_vad_manager.py", line 30, in _add_vad_instance self._instance_map[key] = AudioVadV2(self._vad_config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/lavrock/work/ai_call/.venv/lib/python3.12/site-packages/agora/rtc/voice_detection.py", line 60, in __init__ self._start_size = self._vad_configure.pre_start_recognize_count + self._vad_configure.start_recognize_count ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ AttributeError: 'dict' object has no attribute 'pre_start_recognize_count' 2025-04-27 11:12:53,800 - agora_realtime_ai_api.rtc - INFO - User left: <agora.rtc.rtc_connection.RTCConnection object at 0x16b5a8350> 4034947076 0
AudioVadConfigV2 传这个对象也不知道该怎么配置？
请问我怎么才能在Python SDK 开启vad？