feat: Add mode enum (VERBATIM, SMART) to AudioTranscriptionConfig and TranscriptionConfig.

PiperOrigin-RevId: 964981788

Author: google-genai-bot

google/genai/_gaos/resources/interactions/__init__.py

@@ -97,6 +97,7 @@
 from ...types.interactions.toolchoiceconfig import ToolChoiceConfig
 from ...types.interactions.toolchoicetype import ToolChoiceType
 from ...types.interactions.transcriptionconfig import TranscriptionConfig
+from ...types.interactions.transcriptionmode import TranscriptionMode
 from ...types.interactions.urlcitation import URLCitation
 from ...types.interactions.urlcontextcallarguments import URLContextCallArguments
 from ...types.interactions.urlcontextcallstep import URLContextCallStep
@@ -201,6 +202,7 @@
     "ToolChoiceConfig",
     "ToolChoiceType",
     "TranscriptionConfig",
+    "TranscriptionMode",
     "URLCitation",
     "URLContextCallArguments",
     "URLContextCallStep",

google/genai/_gaos/types/interactions/__init__.py

@@ -366,6 +366,7 @@
     from .toolchoiceconfig import ToolChoiceConfig, ToolChoiceConfigParam
     from .toolchoicetype import ToolChoiceType
     from .transcriptionconfig import TranscriptionConfig, TranscriptionConfigParam
+    from .transcriptionmode import TranscriptionMode
     from .urlcitation import URLCitation, URLCitationParam
     from .urlcontext import URLContext, URLContextParam
     from .urlcontextcallarguments import (
@@ -732,6 +733,7 @@
     "ToolParam",
     "TranscriptionConfig",
     "TranscriptionConfigParam",
+    "TranscriptionMode",
     "Transform",
     "TransformParam",
     "URLCitation",
@@ -1117,6 +1119,7 @@
     "ToolChoiceType": ".toolchoicetype",
     "TranscriptionConfig": ".transcriptionconfig",
     "TranscriptionConfigParam": ".transcriptionconfig",
+    "TranscriptionMode": ".transcriptionmode",
     "URLCitation": ".urlcitation",
     "URLCitationParam": ".urlcitation",
     "URLContext": ".urlcontext",

google/genai/_gaos/types/interactions/interaction.py

@@ -45,7 +45,7 @@
 from .webhookconfig import WebhookConfig, WebhookConfigParam
 from functools import partial
 import pydantic
-from pydantic import ConfigDict, model_serializer, model_validator
+from pydantic import ConfigDict, model_serializer
 from pydantic.functional_validators import BeforeValidator
 from typing import Any, Dict, List, Literal, Optional, Union
 from typing_extensions import Annotated, NotRequired, TypeAliasType, TypedDict
@@ -387,7 +387,7 @@ def _maybe_coerce_outputs(cls, data: Any) -> Any:
         coerced["steps"] = [{"type": "model_output", "content": outputs}]
         return coerced
 
-    @model_validator(mode="before")
+    @pydantic.model_validator(mode="before")
     @classmethod
     def _coerce_outputs_to_steps(cls, data: Any) -> Any:
         return cls._maybe_coerce_outputs(data)
@@ -396,9 +396,11 @@ def _coerce_outputs_to_steps(cls, data: Any) -> Any:
     def model_construct(cls, _fields_set=None, **values):
         # Coerce legacy lyria ``outputs`` -> ``steps`` here as well: validators
         # do not run on model_construct (used by deferred SSE parsing).
-        return super().model_construct(_fields_set, **cls._maybe_coerce_outputs(values))
+        return pydantic.BaseModel.model_construct(
+            _fields_set, **cls._maybe_coerce_outputs(values)
+        )
 
-    @model_validator(mode="after")
+    @pydantic.model_validator(mode="after")
     def _populate_output_helpers(self):
         steps = self.steps if isinstance(self.steps, list) else []
 

google/genai/_gaos/types/interactions/transcriptionconfig.py

@@ -18,6 +18,7 @@
 
 from __future__ import annotations
 from .. import BaseModel, UNSET_SENTINEL
+from .transcriptionmode import TranscriptionMode
 import pydantic
 from pydantic import model_serializer
 from typing import List, Optional
@@ -39,6 +40,11 @@ class TranscriptionConfigParam(TypedDict):
     r"""Optional. BCP-47 language codes providing hints about the languages present in the
     audio. If omitted or empty, defaults to automatic language detection.
     """
+    mode: NotRequired[TranscriptionMode]
+    r"""Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If
+    unspecified, defaults to `VERBATIM` transcription. Mutually exclusive with
+    `timestamp_granularities` and `diarization_mode`.
+    """
     timestamp_granularities: NotRequired[List[str]]
     r"""Optional. The granularity of timestamps to include in the transcription output.
     Supported values: \"word\". If empty, no timestamps are generated.
@@ -69,6 +75,12 @@ class TranscriptionConfig(BaseModel):
     audio. If omitted or empty, defaults to automatic language detection.
     """
 
+    mode: Optional[TranscriptionMode] = None
+    r"""Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If
+    unspecified, defaults to `VERBATIM` transcription. Mutually exclusive with
+    `timestamp_granularities` and `diarization_mode`.
+    """
+
     timestamp_granularities: Optional[List[str]] = None
     r"""Optional. The granularity of timestamps to include in the transcription output.
     Supported values: \"word\". If empty, no timestamps are generated.
@@ -82,6 +94,7 @@ def serialize_model(self, handler):
                 "custom_vocabulary",
                 "diarization_mode",
                 "language_codes",
+                "mode",
                 "timestamp_granularities",
             ]
         )

google/genai/_gaos/types/interactions/transcriptionmode.py

@@ -0,0 +1,34 @@
+# Copyright 2026 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# pyformat: disable
+
+"""Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT."""
+
+from __future__ import annotations
+from .. import UnrecognizedStr
+from typing import Literal, Union
+
+
+TranscriptionMode = Union[
+    Literal[
+        "verbatim",
+        "smart",
+    ],
+    UnrecognizedStr,
+]
+r"""Configures transcription mode. Supported values: `VERBATIM`, `SMART`. If
+unspecified, defaults to `VERBATIM` transcription. Mutually exclusive with
+`timestamp_granularities` and `diarization_mode`.
+"""

google/genai/types.py

@@ -1358,6 +1358,17 @@ 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."""
 
@@ -6369,6 +6380,17 @@ class AudioTranscriptionConfig(_common.BaseModel):
       description="""Configures speaker diarization.
       """,
   )
+  mode: Optional[AudioTranscriptionConfigMode] = Field(
+      default=None,
+      description="""Optional. Transcription mode.
+
+      When set to `SMART`, 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). Incompatible with
+      `word_timestamp` and `diarization`.
+      """,
+  )
 
 
 class AudioTranscriptionConfigDict(TypedDict, total=False):
@@ -6397,6 +6419,16 @@ class AudioTranscriptionConfigDict(TypedDict, total=False):
   """Configures speaker diarization.
       """
 
+  mode: Optional[AudioTranscriptionConfigMode]
+  """Optional. Transcription mode.
+
+      When set to `SMART`, 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). Incompatible with
+      `word_timestamp` and `diarization`.
+      """
+
 
 AudioTranscriptionConfigOrDict = Union[
     AudioTranscriptionConfig, AudioTranscriptionConfigDict