diff --git a/google/genai/_live_converters.py b/google/genai/_live_converters.py
index e3a242587..33e370565 100644
--- a/google/genai/_live_converters.py
+++ b/google/genai/_live_converters.py
@@ -292,7 +292,13 @@ def _GenerationConfig_to_vertex(
     setv(to_object, ['seed'], getv(from_object, ['seed']))
 
   if getv(from_object, ['speech_config']) is not None:
-    setv(to_object, ['speechConfig'], getv(from_object, ['speech_config']))
+    setv(
+        to_object,
+        ['speechConfig'],
+        _SpeechConfig_to_vertex(
+            getv(from_object, ['speech_config']), to_object
+        ),
+    )
 
   if getv(from_object, ['stop_sequences']) is not None:
     setv(to_object, ['stopSequences'], getv(from_object, ['stop_sequences']))
@@ -959,7 +965,10 @@ def _LiveConnectConfig_to_vertex(
     setv(
         parent_object,
         ['setup', 'generationConfig', 'speechConfig'],
-        t.t_live_speech_config(getv(from_object, ['speech_config'])),
+        _SpeechConfig_to_vertex(
+            t.t_live_speech_config(getv(from_object, ['speech_config'])),
+            to_object,
+        ),
     )
 
   if getv(from_object, ['thinking_config']) is not None:
@@ -1391,6 +1400,24 @@ def _LiveServerMessage_from_vertex(
   return to_object
 
 
+def _MultiSpeakerVoiceConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['speaker_voice_configs']) is not None:
+    setv(
+        to_object,
+        ['speakerVoiceConfigs'],
+        [
+            _SpeakerVoiceConfig_to_vertex(item, to_object)
+            for item in getv(from_object, ['speaker_voice_configs'])
+        ],
+    )
+
+  return to_object
+
+
 def _Part_to_mldev(
     from_object: Union[dict[str, Any], object],
     parent_object: Optional[dict[str, Any]] = None,
@@ -1531,6 +1558,32 @@ def _Part_to_vertex(
   return to_object
 
 
+def _ReplicatedVoiceConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['mime_type']) is not None:
+    setv(to_object, ['mimeType'], getv(from_object, ['mime_type']))
+
+  if getv(from_object, ['voice_sample_audio']) is not None:
+    setv(
+        to_object,
+        ['voiceSampleAudio'],
+        getv(from_object, ['voice_sample_audio']),
+    )
+
+  if getv(from_object, ['consent_audio']) is not None:
+    raise ValueError('consent_audio parameter is not supported in Vertex AI.')
+
+  if getv(from_object, ['voice_consent_signature']) is not None:
+    raise ValueError(
+        'voice_consent_signature parameter is not supported in Vertex AI.'
+    )
+
+  return to_object
+
+
 def _SessionResumptionConfig_to_mldev(
     from_object: Union[dict[str, Any], object],
     parent_object: Optional[dict[str, Any]] = None,
@@ -1545,6 +1598,51 @@ def _SessionResumptionConfig_to_mldev(
   return to_object
 
 
+def _SpeakerVoiceConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['speaker']) is not None:
+    setv(to_object, ['speaker'], getv(from_object, ['speaker']))
+
+  if getv(from_object, ['voice_config']) is not None:
+    setv(
+        to_object,
+        ['voiceConfig'],
+        _VoiceConfig_to_vertex(getv(from_object, ['voice_config']), to_object),
+    )
+
+  return to_object
+
+
+def _SpeechConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['voice_config']) is not None:
+    setv(
+        to_object,
+        ['voiceConfig'],
+        _VoiceConfig_to_vertex(getv(from_object, ['voice_config']), to_object),
+    )
+
+  if getv(from_object, ['language_code']) is not None:
+    setv(to_object, ['languageCode'], getv(from_object, ['language_code']))
+
+  if getv(from_object, ['multi_speaker_voice_config']) is not None:
+    setv(
+        to_object,
+        ['multiSpeakerVoiceConfig'],
+        _MultiSpeakerVoiceConfig_to_vertex(
+            getv(from_object, ['multi_speaker_voice_config']), to_object
+        ),
+    )
+
+  return to_object
+
+
 def _Tool_to_mldev(
     from_object: Union[dict[str, Any], object],
     parent_object: Optional[dict[str, Any]] = None,
@@ -1775,3 +1873,27 @@ def _VoiceActivity_from_vertex(
     setv(to_object, ['voice_activity_type'], getv(from_object, ['type']))
 
   return to_object
+
+
+def _VoiceConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['replicated_voice_config']) is not None:
+    setv(
+        to_object,
+        ['replicatedVoiceConfig'],
+        _ReplicatedVoiceConfig_to_vertex(
+            getv(from_object, ['replicated_voice_config']), to_object
+        ),
+    )
+
+  if getv(from_object, ['prebuilt_voice_config']) is not None:
+    setv(
+        to_object,
+        ['prebuiltVoiceConfig'],
+        getv(from_object, ['prebuilt_voice_config']),
+    )
+
+  return to_object
diff --git a/google/genai/live.py b/google/genai/live.py
index 93953a02f..b35e918ce 100644
--- a/google/genai/live.py
+++ b/google/genai/live.py
@@ -92,10 +92,12 @@ def __init__(
       api_client: BaseApiClient,
       websocket: ClientConnection,
       session_id: Optional[str] = None,
+      setup_complete: Optional[types.LiveServerSetupComplete] = None,
   ):
     self._api_client = api_client
     self._ws = websocket
     self.session_id = session_id
+    self.setup_complete = setup_complete
 
   async def send(
       self,
@@ -1131,12 +1133,15 @@ async def connect(
       )
       if setup_response.setup_complete:
         session_id = setup_response.setup_complete.session_id
+        setup_complete = setup_response.setup_complete
       else:
         session_id = None
+        setup_complete = None
       yield AsyncSession(
           api_client=self._api_client,
           websocket=ws,
           session_id=session_id,
+          setup_complete=setup_complete,
       )
 
 
diff --git a/google/genai/models.py b/google/genai/models.py
index c01e8d686..c932b729c 100644
--- a/google/genai/models.py
+++ b/google/genai/models.py
@@ -1468,7 +1468,11 @@ def _GenerateContentConfig_to_vertex(
     setv(
         to_object,
         ['speechConfig'],
-        t.t_speech_config(getv(from_object, ['speech_config'])),
+        _SpeechConfig_to_vertex(
+            t.t_speech_config(getv(from_object, ['speech_config'])),
+            to_object,
+            root_object,
+        ),
     )
 
   if getv(from_object, ['audio_timestamp']) is not None:
@@ -2735,7 +2739,13 @@ def _GenerationConfig_to_vertex(
     setv(to_object, ['seed'], getv(from_object, ['seed']))
 
   if getv(from_object, ['speech_config']) is not None:
-    setv(to_object, ['speechConfig'], getv(from_object, ['speech_config']))
+    setv(
+        to_object,
+        ['speechConfig'],
+        _SpeechConfig_to_vertex(
+            getv(from_object, ['speech_config']), to_object, root_object
+        ),
+    )
 
   if getv(from_object, ['stop_sequences']) is not None:
     setv(to_object, ['stopSequences'], getv(from_object, ['stop_sequences']))
@@ -3301,6 +3311,25 @@ def _Model_from_vertex(
   return to_object
 
 
+def _MultiSpeakerVoiceConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+    root_object: Optional[Union[dict[str, Any], object]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['speaker_voice_configs']) is not None:
+    setv(
+        to_object,
+        ['speakerVoiceConfigs'],
+        [
+            _SpeakerVoiceConfig_to_vertex(item, to_object, root_object)
+            for item in getv(from_object, ['speaker_voice_configs'])
+        ],
+    )
+
+  return to_object
+
+
 def _Part_to_mldev(
     from_object: Union[dict[str, Any], object],
     parent_object: Optional[dict[str, Any]] = None,
@@ -3681,6 +3710,33 @@ def _ReferenceImageAPI_to_vertex(
   return to_object
 
 
+def _ReplicatedVoiceConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+    root_object: Optional[Union[dict[str, Any], object]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['mime_type']) is not None:
+    setv(to_object, ['mimeType'], getv(from_object, ['mime_type']))
+
+  if getv(from_object, ['voice_sample_audio']) is not None:
+    setv(
+        to_object,
+        ['voiceSampleAudio'],
+        getv(from_object, ['voice_sample_audio']),
+    )
+
+  if getv(from_object, ['consent_audio']) is not None:
+    raise ValueError('consent_audio parameter is not supported in Vertex AI.')
+
+  if getv(from_object, ['voice_consent_signature']) is not None:
+    raise ValueError(
+        'voice_consent_signature parameter is not supported in Vertex AI.'
+    )
+
+  return to_object
+
+
 def _SafetyAttributes_from_mldev(
     from_object: Union[dict[str, Any], object],
     parent_object: Optional[dict[str, Any]] = None,
@@ -3883,6 +3939,59 @@ def _SegmentImageSource_to_vertex(
   return to_object
 
 
+def _SpeakerVoiceConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+    root_object: Optional[Union[dict[str, Any], object]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['speaker']) is not None:
+    setv(to_object, ['speaker'], getv(from_object, ['speaker']))
+
+  if getv(from_object, ['voice_config']) is not None:
+    setv(
+        to_object,
+        ['voiceConfig'],
+        _VoiceConfig_to_vertex(
+            getv(from_object, ['voice_config']), to_object, root_object
+        ),
+    )
+
+  return to_object
+
+
+def _SpeechConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+    root_object: Optional[Union[dict[str, Any], object]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['voice_config']) is not None:
+    setv(
+        to_object,
+        ['voiceConfig'],
+        _VoiceConfig_to_vertex(
+            getv(from_object, ['voice_config']), to_object, root_object
+        ),
+    )
+
+  if getv(from_object, ['language_code']) is not None:
+    setv(to_object, ['languageCode'], getv(from_object, ['language_code']))
+
+  if getv(from_object, ['multi_speaker_voice_config']) is not None:
+    setv(
+        to_object,
+        ['multiSpeakerVoiceConfig'],
+        _MultiSpeakerVoiceConfig_to_vertex(
+            getv(from_object, ['multi_speaker_voice_config']),
+            to_object,
+            root_object,
+        ),
+    )
+
+  return to_object
+
+
 def _ToolConfig_to_mldev(
     from_object: Union[dict[str, Any], object],
     parent_object: Optional[dict[str, Any]] = None,
@@ -4481,6 +4590,33 @@ def _Video_to_vertex(
   return to_object
 
 
+def _VoiceConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+    root_object: Optional[Union[dict[str, Any], object]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['replicated_voice_config']) is not None:
+    setv(
+        to_object,
+        ['replicatedVoiceConfig'],
+        _ReplicatedVoiceConfig_to_vertex(
+            getv(from_object, ['replicated_voice_config']),
+            to_object,
+            root_object,
+        ),
+    )
+
+  if getv(from_object, ['prebuilt_voice_config']) is not None:
+    setv(
+        to_object,
+        ['prebuiltVoiceConfig'],
+        getv(from_object, ['prebuilt_voice_config']),
+    )
+
+  return to_object
+
+
 class Models(_api_module.BaseModule):
 
   def _generate_content(
diff --git a/google/genai/tests/live/test_live.py b/google/genai/tests/live/test_live.py
index 2d59ce900..513a90a3f 100644
--- a/google/genai/tests/live/test_live.py
+++ b/google/genai/tests/live/test_live.py
@@ -2154,3 +2154,106 @@ async def mock_connect(uri, additional_headers=None, **kwargs):
   assert capture['headers']['x-goog-api-key'] == 'TEST_API_KEY'
   assert 'BidiGenerateContent' in capture['uri']
 
+
+@pytest.mark.parametrize('vertexai', [True, False])
+@pytest.mark.asyncio
+async def test_async_session_setup_complete_with_voice_consent_signature(
+    vertexai,
+):
+  mock_ws = AsyncMock()
+  mock_ws.send = AsyncMock()
+  mock_ws.recv = AsyncMock(
+      return_value=(
+          b'{"setupComplete": {"sessionId": "test_session_id",'
+          b' "voiceConsentSignature": {"signature": "test_sig_abc123"}}}'
+      )
+  )
+  mock_ws.close = AsyncMock()
+
+  mock_google_auth_default = Mock(return_value=(None, None))
+  mock_creds = Mock(token='test_token')
+  mock_google_auth_default.return_value = (mock_creds, None)
+
+  @contextlib.asynccontextmanager
+  async def mock_connect(uri, additional_headers=None, **kwargs):
+    yield mock_ws
+
+  @patch('google.auth.default', new=mock_google_auth_default)
+  @patch.object(live, 'ws_connect', new=mock_connect)
+  async def _test_connect():
+    live_module = live.AsyncLive(mock_api_client(vertexai=vertexai))
+    async with live_module.connect(model='test_model') as session:
+      assert session.setup_complete is not None
+      assert session.setup_complete.session_id == 'test_session_id'
+      assert session.setup_complete.voice_consent_signature is not None
+      assert (
+          session.setup_complete.voice_consent_signature.signature
+          == 'test_sig_abc123'
+      )
+
+  await _test_connect()
+
+
+@pytest.mark.parametrize('vertexai', [False])
+@pytest.mark.asyncio
+async def test_bidi_setup_replicated_voice_config_with_consent(vertexai):
+  config = types.LiveConnectConfig(
+      response_modalities=['AUDIO'],
+      speech_config=types.SpeechConfig(
+          voice_config=types.VoiceConfig(
+              replicated_voice_config=types.ReplicatedVoiceConfig(
+                  mime_type='audio/wav',
+                  voice_sample_audio=b'fake_audio_data',
+                  consent_audio=b'fake_consent_data',
+              )
+          )
+      ),
+  )
+  result = await get_connect_message(
+      mock_api_client(vertexai=vertexai), model='test_model', config=config
+  )
+
+  setup = result.get('setup', {})
+  gen_config = setup.get('generationConfig', {})
+  speech_config = gen_config.get('speechConfig', {})
+  voice_config = speech_config.get('voice_config', {})
+  replicated = voice_config.get('replicated_voice_config', {})
+
+  assert replicated.get('mime_type') == 'audio/wav'
+  assert replicated.get('voice_sample_audio') is not None
+  assert replicated.get('consent_audio') is not None
+
+  config_with_sig = types.LiveConnectConfig(
+      response_modalities=['AUDIO'],
+      speech_config=types.SpeechConfig(
+          voice_config=types.VoiceConfig(
+              replicated_voice_config=types.ReplicatedVoiceConfig(
+                  mime_type='audio/wav',
+                  voice_sample_audio=b'fake_audio_data',
+                  voice_consent_signature=types.VoiceConsentSignature(
+                      signature='test_sig_abc123'
+                  ),
+              )
+          )
+      ),
+  )
+  result_with_sig = await get_connect_message(
+      mock_api_client(vertexai=vertexai),
+      model='test_model',
+      config=config_with_sig,
+  )
+
+  setup_sig = result_with_sig.get('setup', {})
+  gen_config_sig = setup_sig.get('generationConfig', {})
+  speech_config_sig = gen_config_sig.get('speechConfig', {})
+  voice_config_sig = speech_config_sig.get('voice_config', {})
+  replicated_sig = voice_config_sig.get('replicated_voice_config', {})
+
+  assert replicated_sig.get('mime_type') == 'audio/wav'
+  assert replicated_sig.get('voice_sample_audio') is not None
+  assert replicated_sig.get('voice_consent_signature') is not None
+  assert (
+      replicated_sig['voice_consent_signature'].get('signature')
+      == 'test_sig_abc123'
+  )
+
diff --git a/google/genai/tunings.py b/google/genai/tunings.py
index d6dfb0642..9566ebb43 100644
--- a/google/genai/tunings.py
+++ b/google/genai/tunings.py
@@ -832,7 +832,13 @@ def _GenerationConfig_to_vertex(
     setv(to_object, ['seed'], getv(from_object, ['seed']))
 
   if getv(from_object, ['speech_config']) is not None:
-    setv(to_object, ['speechConfig'], getv(from_object, ['speech_config']))
+    setv(
+        to_object,
+        ['speechConfig'],
+        _SpeechConfig_to_vertex(
+            getv(from_object, ['speech_config']), to_object, root_object
+        ),
+    )
 
   if getv(from_object, ['stop_sequences']) is not None:
     setv(to_object, ['stopSequences'], getv(from_object, ['stop_sequences']))
@@ -1013,6 +1019,105 @@ def _ListTuningJobsResponse_from_vertex(
   return to_object
 
 
+def _MultiSpeakerVoiceConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+    root_object: Optional[Union[dict[str, Any], object]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['speaker_voice_configs']) is not None:
+    setv(
+        to_object,
+        ['speakerVoiceConfigs'],
+        [
+            _SpeakerVoiceConfig_to_vertex(item, to_object, root_object)
+            for item in getv(from_object, ['speaker_voice_configs'])
+        ],
+    )
+
+  return to_object
+
+
+def _ReplicatedVoiceConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+    root_object: Optional[Union[dict[str, Any], object]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['mime_type']) is not None:
+    setv(to_object, ['mimeType'], getv(from_object, ['mime_type']))
+
+  if getv(from_object, ['voice_sample_audio']) is not None:
+    setv(
+        to_object,
+        ['voiceSampleAudio'],
+        getv(from_object, ['voice_sample_audio']),
+    )
+
+  if getv(from_object, ['consent_audio']) is not None:
+    raise ValueError('consent_audio parameter is not supported in Vertex AI.')
+
+  if getv(from_object, ['voice_consent_signature']) is not None:
+    raise ValueError(
+        'voice_consent_signature parameter is not supported in Vertex AI.'
+    )
+
+  return to_object
+
+
+def _SpeakerVoiceConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+    root_object: Optional[Union[dict[str, Any], object]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['speaker']) is not None:
+    setv(to_object, ['speaker'], getv(from_object, ['speaker']))
+
+  if getv(from_object, ['voice_config']) is not None:
+    setv(
+        to_object,
+        ['voiceConfig'],
+        _VoiceConfig_to_vertex(
+            getv(from_object, ['voice_config']), to_object, root_object
+        ),
+    )
+
+  return to_object
+
+
+def _SpeechConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+    root_object: Optional[Union[dict[str, Any], object]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['voice_config']) is not None:
+    setv(
+        to_object,
+        ['voiceConfig'],
+        _VoiceConfig_to_vertex(
+            getv(from_object, ['voice_config']), to_object, root_object
+        ),
+    )
+
+  if getv(from_object, ['language_code']) is not None:
+    setv(to_object, ['languageCode'], getv(from_object, ['language_code']))
+
+  if getv(from_object, ['multi_speaker_voice_config']) is not None:
+    setv(
+        to_object,
+        ['multiSpeakerVoiceConfig'],
+        _MultiSpeakerVoiceConfig_to_vertex(
+            getv(from_object, ['multi_speaker_voice_config']),
+            to_object,
+            root_object,
+        ),
+    )
+
+  return to_object
+
+
 def _TunedModel_from_mldev(
     from_object: Union[dict[str, Any], object],
     parent_object: Optional[dict[str, Any]] = None,
@@ -1376,6 +1481,33 @@ def _TuningValidationDataset_to_vertex(
   return to_object
 
 
+def _VoiceConfig_to_vertex(
+    from_object: Union[dict[str, Any], object],
+    parent_object: Optional[dict[str, Any]] = None,
+    root_object: Optional[Union[dict[str, Any], object]] = None,
+) -> dict[str, Any]:
+  to_object: dict[str, Any] = {}
+  if getv(from_object, ['replicated_voice_config']) is not None:
+    setv(
+        to_object,
+        ['replicatedVoiceConfig'],
+        _ReplicatedVoiceConfig_to_vertex(
+            getv(from_object, ['replicated_voice_config']),
+            to_object,
+            root_object,
+        ),
+    )
+
+  if getv(from_object, ['prebuilt_voice_config']) is not None:
+    setv(
+        to_object,
+        ['prebuiltVoiceConfig'],
+        getv(from_object, ['prebuilt_voice_config']),
+    )
+
+  return to_object
+
+
 class Tunings(_api_module.BaseModule):
 
   def _get(
diff --git a/google/genai/types.py b/google/genai/types.py
index 41da322c2..35f5e5ffb 100644
--- a/google/genai/types.py
+++ b/google/genai/types.py
@@ -4956,32 +4956,85 @@ class ToolConfigDict(TypedDict, total=False):
 ToolConfigOrDict = Union[ToolConfig, ToolConfigDict]
 
 
+class VoiceConsentSignature(_common.BaseModel):
+  """The signature of the voice consent check."""
+
+  signature: Optional[str] = Field(
+      default=None,
+      description="""The signature string.
+      """,
+  )
+
+
+class VoiceConsentSignatureDict(TypedDict, total=False):
+  """The signature of the voice consent check."""
+
+  signature: Optional[str]
+  """The signature string.
+      """
+
+
+VoiceConsentSignatureOrDict = Union[
+    VoiceConsentSignature, VoiceConsentSignatureDict
+]
+
+
 class ReplicatedVoiceConfig(_common.BaseModel):
-  """ReplicatedVoiceConfig is used to configure replicated voice."""
+  """The configuration for the replicated voice to use."""
 
   mime_type: Optional[str] = Field(
       default=None,
-      description="""The mime type of the replicated voice.
+      description="""The mimetype of the voice sample. The only currently supported
+      value is `audio/wav`. This represents 16-bit signed little-endian wav
+      data, with a 24kHz sampling rate.
       """,
   )
   voice_sample_audio: Optional[bytes] = Field(
       default=None,
-      description="""The sample audio of the replicated voice.
+      description="""The sample of the custom voice.
       """,
   )
+  consent_audio: Optional[bytes] = Field(
+      default=None,
+      description="""Recorded consent verifying ownership of the voice. This
+      represents 16-bit signed little-endian wav data, with a 24kHz sampling
+      rate.""",
+  )
+  voice_consent_signature: Optional[VoiceConsentSignature] = Field(
+      default=None,
+      description="""Signature of a previously verified consent audio. This should be
+      populated with a signature generated by the server for a previous
+      request containing the consent_audio field. When provided, the
+      signature is verified instead of the consent_audio field to reduce
+      latency. Requests will fail if the signature is invalid or expired.""",
+  )
 
 
 class ReplicatedVoiceConfigDict(TypedDict, total=False):
-  """ReplicatedVoiceConfig is used to configure replicated voice."""
+  """The configuration for the replicated voice to use."""
 
   mime_type: Optional[str]
-  """The mime type of the replicated voice.
+  """The mimetype of the voice sample. The only currently supported
+      value is `audio/wav`. This represents 16-bit signed little-endian wav
+      data, with a 24kHz sampling rate.
       """
 
   voice_sample_audio: Optional[bytes]
-  """The sample audio of the replicated voice.
+  """The sample of the custom voice.
       """
 
+  consent_audio: Optional[bytes]
+  """Recorded consent verifying ownership of the voice. This
+      represents 16-bit signed little-endian wav data, with a 24kHz sampling
+      rate."""
+
+  voice_consent_signature: Optional[VoiceConsentSignatureDict]
+  """Signature of a previously verified consent audio. This should be
+      populated with a signature generated by the server for a previous
+      request containing the consent_audio field. When provided, the
+      signature is verified instead of the consent_audio field to reduce
+      latency. Requests will fail if the signature is invalid or expired."""
+
 
 ReplicatedVoiceConfigOrDict = Union[
     ReplicatedVoiceConfig, ReplicatedVoiceConfigDict
@@ -5007,10 +5060,13 @@ class PrebuiltVoiceConfigDict(TypedDict, total=False):
 
 
 class VoiceConfig(_common.BaseModel):
+  """The configuration for the voice to use."""
 
   replicated_voice_config: Optional[ReplicatedVoiceConfig] = Field(
       default=None,
-      description="""If true, the model will use a replicated voice for the response.""",
+      description="""The configuration for a replicated voice, which is a clone of a
+      user's voice that can be used for speech synthesis. If this is unset, a
+      default voice is used.""",
   )
   prebuilt_voice_config: Optional[PrebuiltVoiceConfig] = Field(
       default=None, description="""The configuration for a prebuilt voice."""
@@ -5018,9 +5074,12 @@ class VoiceConfig(_common.BaseModel):
 
 
 class VoiceConfigDict(TypedDict, total=False):
+  """The configuration for the voice to use."""
 
   replicated_voice_config: Optional[ReplicatedVoiceConfigDict]
-  """If true, the model will use a replicated voice for the response."""
+  """The configuration for a replicated voice, which is a clone of a
+      user's voice that can be used for speech synthesis. If this is unset, a
+      default voice is used."""
 
   prebuilt_voice_config: Optional[PrebuiltVoiceConfigDict]
   """The configuration for a prebuilt voice."""
@@ -5077,10 +5136,11 @@ class MultiSpeakerVoiceConfigDict(TypedDict, total=False):
 
 
 class SpeechConfig(_common.BaseModel):
+  """Config for speech generation and transcription."""
 
   voice_config: Optional[VoiceConfig] = Field(
       default=None,
-      description="""Configuration for the voice of the response.""",
+      description="""The configuration in case of single-voice output.""",
   )
   language_code: Optional[str] = Field(
       default=None,
@@ -5093,9 +5153,10 @@ class SpeechConfig(_common.BaseModel):
 
 
 class SpeechConfigDict(TypedDict, total=False):
+  """Config for speech generation and transcription."""
 
   voice_config: Optional[VoiceConfigDict]
-  """Configuration for the voice of the response."""
+  """The configuration in case of single-voice output."""
 
   language_code: Optional[str]
   """Optional. The language code (ISO 639-1) for the speech synthesis."""
@@ -17789,6 +17850,13 @@ class LiveServerSetupComplete(_common.BaseModel):
   session_id: Optional[str] = Field(
       default=None, description="""The session id of the live session."""
   )
+  voice_consent_signature: Optional[VoiceConsentSignature] = Field(
+      default=None,
+      description="""Signature of the verified consent audio. This is populated when the
+      request has a ReplicatedVoiceConfig with consent_audio set, if the consent
+      verification was successful. This may be used in a subsequent request
+      instead of the consent_audio to verify the same consent.""",
+  )
 
 
 class LiveServerSetupCompleteDict(TypedDict, total=False):
@@ -17797,6 +17865,12 @@ class LiveServerSetupCompleteDict(TypedDict, total=False):
   session_id: Optional[str]
   """The session id of the live session."""
 
+  voice_consent_signature: Optional[VoiceConsentSignatureDict]
+  """Signature of the verified consent audio. This is populated when the
+      request has a ReplicatedVoiceConfig with consent_audio set, if the consent
+      verification was successful. This may be used in a subsequent request
+      instead of the consent_audio to verify the same consent."""
+
 
 LiveServerSetupCompleteOrDict = Union[
     LiveServerSetupComplete, LiveServerSetupCompleteDict