feat: Add AudioTranscriptionConfigMode

PiperOrigin-RevId: 970192656

Author: MarkDaoust

google/genai/tests/models/test_generate_content.py

@@ -617,6 +617,26 @@ class InstrumentEnum(Enum):
             ),
         ),
     ),
+    pytest_helper.TestTableItem(
+        name='test_audio_transcription_config_mode',
+        parameters=types._GenerateContentParameters(
+            model='gemini-2.5-flash-preview-tts',
+            contents=t.t_contents('Produce a speech response saying "Cheese"'),
+            config=types.GenerateContentConfig(
+                response_modalities=['audio'],
+                speech_config=types.SpeechConfig(
+                    voice_config=types.VoiceConfig(
+                        prebuilt_voice_config=types.PrebuiltVoiceConfig(
+                            voice_name='charon'
+                        )
+                    )
+                ),
+                audio_transcription_config=types.AudioTranscriptionConfig(
+                    mode='SMART',
+                ),
+            ),
+        ),
+    ),
 ]
 
 pytestmark = pytest_helper.setup(

google/genai/types.py

@@ -1373,6 +1373,17 @@ class VoiceActivityType(_common.CaseInSensitiveEnum):
   """End of sentence signal."""
 
 
+class AudioTranscriptionConfigMode(_common.CaseInSensitiveEnum):
+  """Transcription mode."""
+
+  MODE_UNSPECIFIED = 'MODE_UNSPECIFIED'
+  """Unspecified transcription mode."""
+  VERBATIM = 'VERBATIM'
+  """Verbatim transcription mode."""
+  SMART = 'SMART'
+  """Smart transcription mode."""
+
+
 class StartSensitivity(_common.CaseInSensitiveEnum):
   """Start of speech sensitivity."""
 
@@ -6391,6 +6402,10 @@ class AudioTranscriptionConfig(_common.BaseModel):
       description="""Configures speaker diarization.
       """,
   )
+  mode: Optional[AudioTranscriptionConfigMode] = Field(
+      default=None,
+      description="""Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`.""",
+  )
 
 
 class AudioTranscriptionConfigDict(TypedDict, total=False):
@@ -6419,6 +6434,9 @@ class AudioTranscriptionConfigDict(TypedDict, total=False):
   """Configures speaker diarization.
       """
 
+  mode: Optional[AudioTranscriptionConfigMode]
+  """Optional. Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If unspecified, defaults to `VERBATIM` transcription. In `SMART` mode, the model performs disfluency removal (eliminating filler words, repetitions, and false starts), light grammatical cleanup, automatic formatting (paragraphs, bullet points, numbered lists), and minor user edits (inline self-corrections). Timestamps and diarization are incompatible with mode `SMART`."""
+
 
 AudioTranscriptionConfigOrDict = Union[
     AudioTranscriptionConfig, AudioTranscriptionConfigDict