chore: Tests for ReplicatedVoiceConfig

PiperOrigin-RevId: 841593648

Author: MarkDaoust

google/genai/_live_converters.py

@@ -1265,12 +1265,12 @@ def _SpeechConfig_to_vertex(
     parent_object: Optional[dict[str, Any]] = None,
 ) -> dict[str, Any]:
   to_object: dict[str, Any] = {}
-  if getv(from_object, ['language_code']) is not None:
-    setv(to_object, ['languageCode'], getv(from_object, ['language_code']))
-
   if getv(from_object, ['voice_config']) is not None:
     setv(to_object, ['voiceConfig'], getv(from_object, ['voice_config']))
 
+  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:
     raise ValueError(
         'multi_speaker_voice_config parameter is not supported in Vertex AI.'

google/genai/models.py

@@ -3420,12 +3420,12 @@ def _SpeechConfig_to_vertex(
     parent_object: Optional[dict[str, Any]] = None,
 ) -> dict[str, Any]:
   to_object: dict[str, Any] = {}
-  if getv(from_object, ['language_code']) is not None:
-    setv(to_object, ['languageCode'], getv(from_object, ['language_code']))
-
   if getv(from_object, ['voice_config']) is not None:
     setv(to_object, ['voiceConfig'], getv(from_object, ['voice_config']))
 
+  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:
     raise ValueError(
         'multi_speaker_voice_config parameter is not supported in Vertex AI.'

google/genai/tests/data/voice_sample.wav

@@ -808,6 +808,44 @@ async def test_bidi_setup_error_if_multispeaker_voice_config(vertexai):
     )
 
 
+@pytest.mark.parametrize('vertexai', [True, False])
+@pytest.mark.asyncioasync
+async def test_replicated_voice_config(vertexai):
+  # Config is a dict
+  config_dict = {
+      'speech_config': {
+          'voice_config': {
+              'replicated_voice_config': {
+                  'mime_type': 'audio/pcm',
+                  ' voice_sample_audio ': bytes([0, 0, 0]),
+              },
+          },
+      },
+  }
+  result = await get_connect_message(
+      mock_api_client(vertexai=vertexai),
+      model='test_model',
+      config=config_dict,
+  )
+  assert (
+      result['setup']['generationConfig']['speechConfig']['voiceConfig'][
+          'replicatedVoiceConfig'
+      ]
+      == 'AAAA'
+  )
+
+
+@pytest.mark.parametrize('vertexai', [True, False])
+@pytest.mark.asyncioasync
+async def test_explicit_vad(vertexai):
+  # Config is a dict
+  config_dict = {'explicit_vad_signal': True}
+  result = await get_connect_message(
+      mock_api_client(vertexai=vertexai), model='test_model', config=config_dict
+  )
+  assert result['setup']['explicitVadSignal'] == True
+
+
 @pytest.mark.parametrize('vertexai', [True, False])
 @pytest.mark.asyncio
 async def test_bidi_setup_to_api_with_system_instruction_as_content_type(

google/genai/tests/live/test_live.py

@@ -13,9 +13,8 @@
 # limitations under the License.
 #
 
-import base64
-import enum
 import os
+import pathlib
 
 from pydantic import BaseModel, ValidationError, Field, ConfigDict
 from typing import Literal, List, Optional, Union, Set
@@ -30,16 +29,16 @@
 from .. import pytest_helper
 from enum import Enum
 
-IMAGE_PNG_FILE_PATH = os.path.abspath(
-    os.path.join(os.path.dirname(__file__), '../data/google.png')
-)
-
 GEMINI_FLASH_LATEST = 'gemini-2.5-flash'
 GEMINI_FLASH_2_0 = 'gemini-2.0-flash-001'
 GEMINI_FLASH_IMAGE_LATEST = 'gemini-2.5-flash-image'
 
-with open(IMAGE_PNG_FILE_PATH, 'rb') as image_file:
-  image_bytes = image_file.read()
+IMAGE_PNG_FILE_PATH = pathlib.Path(__file__).parent / '../data/google.png'
+image_bytes = IMAGE_PNG_FILE_PATH.read_bytes()
+
+AUDIO_WAV_FILE_PATH = pathlib.Path(__file__).parent / '../data/voice_sample.wav'
+audio_bytes = AUDIO_WAV_FILE_PATH.read_bytes()
+
 
 safety_settings_with_method = [
     {
@@ -1964,6 +1963,27 @@ def test_schema_with_any_of(client):
   assert 'type' in response.parsed['fruit'][0]
 
 
+def test_replicated_voice_config(client):
+  with pytest_helper.exception_if_vertex(client, errors.ClientError):
+      client.models.generate_content(
+          model='gemini-2.5-flash-preview-tts-voice-replication-rev22-2025-10-28',
+          contents=t.t_contents(
+              'Produce a speech response saying "Cheese"'
+          ),
+          config=types.GenerateContentConfig(
+              response_modalities=['audio'],
+              speech_config=types.SpeechConfig(
+                  voice_config=types.VoiceConfig(
+                      replicated_voice_config=types.ReplicatedVoiceConfig(
+                          voice_sample_audio=audio_bytes,
+                          mime_type='audio/wav',
+                      )
+                  )
+              ),
+          ),
+      )
+
+
 def test_json_schema_with_streaming(client):
 
   response = client.models.generate_content_stream(

google/genai/tests/models/test_generate_content.py

@@ -867,12 +867,12 @@ def _SpeechConfig_to_vertex(
     root_object: Optional[Union[dict[str, Any], object]] = None,
 ) -> dict[str, Any]:
   to_object: dict[str, Any] = {}
-  if getv(from_object, ['language_code']) is not None:
-    setv(to_object, ['languageCode'], getv(from_object, ['language_code']))
-
   if getv(from_object, ['voice_config']) is not None:
     setv(to_object, ['voiceConfig'], getv(from_object, ['voice_config']))
 
+  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:
     raise ValueError(
         'multi_speaker_voice_config parameter is not supported in Vertex AI.'