chore: Add tests for generate content transcription.

google/genai/_live_converters.py

@@ -47,6 +47,12 @@ def _AudioTranscriptionConfig_to_mldev(
         getv(from_object, ['adaptation_phrases']),
     )
 
+  if getv(from_object, ['word_timestamp']) is not None:
+    setv(to_object, ['wordTimestamp'], getv(from_object, ['word_timestamp']))
+
+  if getv(from_object, ['diarization']) is not None:
+    setv(to_object, ['diarization'], getv(from_object, ['diarization']))
+
   return to_object
 
 
@@ -409,6 +415,13 @@ def _GenerationConfig_to_vertex(
         ' mode, not in Gemini Enterprise Agent Platform mode.'
     )
 
+  if getv(from_object, ['audio_transcription_config']) is not None:
+    setv(
+        to_object,
+        ['audioTranscriptionConfig'],
+        getv(from_object, ['audio_transcription_config']),
+    )
+
   return to_object
 
 
@@ -1689,6 +1702,13 @@ def _Part_to_mldev(
   if getv(from_object, ['part_metadata']) is not None:
     setv(to_object, ['partMetadata'], getv(from_object, ['part_metadata']))
 
+  if getv(from_object, ['audio_transcription']) is not None:
+    setv(
+        to_object,
+        ['audioTranscription'],
+        getv(from_object, ['audio_transcription']),
+    )
+
   return to_object
 
 
@@ -1770,6 +1790,13 @@ def _Part_to_vertex(
         ' mode, not in Gemini Enterprise Agent Platform mode.'
     )
 
+  if getv(from_object, ['audio_transcription']) is not None:
+    setv(
+        to_object,
+        ['audioTranscription'],
+        getv(from_object, ['audio_transcription']),
+    )
+
   return to_object
 
 

google/genai/_tokens_converters.py

@@ -47,6 +47,12 @@ def _AudioTranscriptionConfig_to_mldev(
         getv(from_object, ['adaptation_phrases']),
     )
 
+  if getv(from_object, ['word_timestamp']) is not None:
+    setv(to_object, ['wordTimestamp'], getv(from_object, ['word_timestamp']))
+
+  if getv(from_object, ['diarization']) is not None:
+    setv(to_object, ['diarization'], getv(from_object, ['diarization']))
+
   return to_object
 
 
@@ -595,6 +601,13 @@ def _Part_to_mldev(
   if getv(from_object, ['part_metadata']) is not None:
     setv(to_object, ['partMetadata'], getv(from_object, ['part_metadata']))
 
+  if getv(from_object, ['audio_transcription']) is not None:
+    setv(
+        to_object,
+        ['audioTranscription'],
+        getv(from_object, ['audio_transcription']),
+    )
+
   return to_object
 
 

google/genai/batches.py

@@ -34,6 +34,39 @@
 logger = logging.getLogger('google_genai.batches')
 
 
+def _AudioTranscriptionConfig_to_mldev(
+    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, ['language_codes']) is not None:
+    raise ValueError(
+        'language_codes parameter is only supported in Gemini Enterprise Agent'
+        ' Platform mode, not in Gemini Developer API mode.'
+    )
+
+  if getv(from_object, ['language_auto']) is not None:
+    setv(to_object, ['languageAuto'], getv(from_object, ['language_auto']))
+
+  if getv(from_object, ['language_hints']) is not None:
+    setv(to_object, ['languageHints'], getv(from_object, ['language_hints']))
+
+  if getv(from_object, ['adaptation_phrases']) is not None:
+    setv(
+        to_object,
+        ['adaptationPhrases'],
+        getv(from_object, ['adaptation_phrases']),
+    )
+
+  if getv(from_object, ['word_timestamp']) is not None:
+    setv(to_object, ['wordTimestamp'], getv(from_object, ['word_timestamp']))
+
+  if getv(from_object, ['diarization']) is not None:
+    setv(to_object, ['diarization'], getv(from_object, ['diarization']))
+
+  return to_object
+
+
 def _AuthConfig_to_mldev(
     from_object: Union[dict[str, Any], object],
     parent_object: Optional[dict[str, Any]] = None,
@@ -1182,6 +1215,15 @@ def _GenerateContentConfig_to_mldev(
   if getv(from_object, ['service_tier']) is not None:
     setv(parent_object, ['serviceTier'], getv(from_object, ['service_tier']))
 
+  if getv(from_object, ['audio_transcription_config']) is not None:
+    setv(
+        to_object,
+        ['audioTranscriptionConfig'],
+        _AudioTranscriptionConfig_to_mldev(
+            getv(from_object, ['audio_transcription_config']), to_object
+        ),
+    )
+
   return to_object
 
 
@@ -1606,6 +1648,13 @@ def _Part_to_mldev(
   if getv(from_object, ['part_metadata']) is not None:
     setv(to_object, ['partMetadata'], getv(from_object, ['part_metadata']))
 
+  if getv(from_object, ['audio_transcription']) is not None:
+    setv(
+        to_object,
+        ['audioTranscription'],
+        getv(from_object, ['audio_transcription']),
+    )
+
   return to_object
 
 

google/genai/caches.py

@@ -789,6 +789,13 @@ def _Part_to_mldev(
   if getv(from_object, ['part_metadata']) is not None:
     setv(to_object, ['partMetadata'], getv(from_object, ['part_metadata']))
 
+  if getv(from_object, ['audio_transcription']) is not None:
+    setv(
+        to_object,
+        ['audioTranscription'],
+        getv(from_object, ['audio_transcription']),
+    )
+
   return to_object
 
 
@@ -870,6 +877,13 @@ def _Part_to_vertex(
         ' mode, not in Gemini Enterprise Agent Platform mode.'
     )
 
+  if getv(from_object, ['audio_transcription']) is not None:
+    setv(
+        to_object,
+        ['audioTranscription'],
+        getv(from_object, ['audio_transcription']),
+    )
+
   return to_object
 
 

google/genai/models.py

@@ -64,6 +64,40 @@ def _VideoGenerationReferenceType_to_mldev_enum_validate(
     )
 
 
+def _AudioTranscriptionConfig_to_mldev(
+    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, ['language_codes']) is not None:
+    raise ValueError(
+        'language_codes parameter is only supported in Gemini Enterprise Agent'
+        ' Platform mode, not in Gemini Developer API mode.'
+    )
+
+  if getv(from_object, ['language_auto']) is not None:
+    setv(to_object, ['languageAuto'], getv(from_object, ['language_auto']))
+
+  if getv(from_object, ['language_hints']) is not None:
+    setv(to_object, ['languageHints'], getv(from_object, ['language_hints']))
+
+  if getv(from_object, ['adaptation_phrases']) is not None:
+    setv(
+        to_object,
+        ['adaptationPhrases'],
+        getv(from_object, ['adaptation_phrases']),
+    )
+
+  if getv(from_object, ['word_timestamp']) is not None:
+    setv(to_object, ['wordTimestamp'], getv(from_object, ['word_timestamp']))
+
+  if getv(from_object, ['diarization']) is not None:
+    setv(to_object, ['diarization'], getv(from_object, ['diarization']))
+
+  return to_object
+
+
 def _AuthConfig_to_mldev(
     from_object: Union[dict[str, Any], object],
     parent_object: Optional[dict[str, Any]] = None,
@@ -1453,6 +1487,17 @@ def _GenerateContentConfig_to_mldev(
   if getv(from_object, ['service_tier']) is not None:
     setv(parent_object, ['serviceTier'], getv(from_object, ['service_tier']))
 
+  if getv(from_object, ['audio_transcription_config']) is not None:
+    setv(
+        to_object,
+        ['audioTranscriptionConfig'],
+        _AudioTranscriptionConfig_to_mldev(
+            getv(from_object, ['audio_transcription_config']),
+            to_object,
+            root_object,
+        ),
+    )
+
   return to_object
 
 
@@ -1643,6 +1688,13 @@ def _GenerateContentConfig_to_vertex(
   if getv(from_object, ['service_tier']) is not None:
     setv(parent_object, ['serviceTier'], getv(from_object, ['service_tier']))
 
+  if getv(from_object, ['audio_transcription_config']) is not None:
+    setv(
+        to_object,
+        ['audioTranscriptionConfig'],
+        getv(from_object, ['audio_transcription_config']),
+    )
+
   return to_object
 
 
@@ -2990,6 +3042,13 @@ def _GenerationConfig_to_vertex(
         ' mode, not in Gemini Enterprise Agent Platform mode.'
     )
 
+  if getv(from_object, ['audio_transcription_config']) is not None:
+    setv(
+        to_object,
+        ['audioTranscriptionConfig'],
+        getv(from_object, ['audio_transcription_config']),
+    )
+
   return to_object
 
 
@@ -3664,6 +3723,13 @@ def _Part_to_mldev(
   if getv(from_object, ['part_metadata']) is not None:
     setv(to_object, ['partMetadata'], getv(from_object, ['part_metadata']))
 
+  if getv(from_object, ['audio_transcription']) is not None:
+    setv(
+        to_object,
+        ['audioTranscription'],
+        getv(from_object, ['audio_transcription']),
+    )
+
   return to_object
 
 
@@ -3746,6 +3812,13 @@ def _Part_to_vertex(
         ' mode, not in Gemini Enterprise Agent Platform mode.'
     )
 
+  if getv(from_object, ['audio_transcription']) is not None:
+    setv(
+        to_object,
+        ['audioTranscription'],
+        getv(from_object, ['audio_transcription']),
+    )
+
   return to_object