feat: Make speech_config a structured object.

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

@@ -162,6 +162,8 @@
     from .generationconfig import (
         GenerationConfig,
         GenerationConfigParam,
+        SpeechConfigUnion,
+        SpeechConfigUnionParam,
         ToolChoice,
         ToolChoiceParam,
     )
@@ -325,6 +327,7 @@
     from .servicetier import ServiceTier
     from .sessionconfig import SessionConfig, SessionConfigParam
     from .source import Source, SourceParam, SourceType
+    from .speakerconfig import SpeakerConfig, SpeakerConfigParam
     from .speechconfig import SpeechConfig, SpeechConfigParam
     from .staticmediaprocessing import StaticMediaProcessing, StaticMediaProcessingParam
     from .status import Status, StatusParam
@@ -676,8 +679,12 @@
     "Source",
     "SourceParam",
     "SourceType",
+    "SpeakerConfig",
+    "SpeakerConfigParam",
     "SpeechConfig",
     "SpeechConfigParam",
+    "SpeechConfigUnion",
+    "SpeechConfigUnionParam",
     "StaticMediaProcessing",
     "StaticMediaProcessingParam",
     "Status",
@@ -915,6 +922,8 @@
     "UnknownFunctionResultSubcontent": ".functionresultsubcontent",
     "GenerationConfig": ".generationconfig",
     "GenerationConfigParam": ".generationconfig",
+    "SpeechConfigUnion": ".generationconfig",
+    "SpeechConfigUnionParam": ".generationconfig",
     "ToolChoice": ".generationconfig",
     "ToolChoiceParam": ".generationconfig",
     "GoogleMaps": ".googlemaps",
@@ -1061,6 +1070,8 @@
     "Source": ".source",
     "SourceParam": ".source",
     "SourceType": ".source",
+    "SpeakerConfig": ".speakerconfig",
+    "SpeakerConfigParam": ".speakerconfig",
     "SpeechConfig": ".speechconfig",
     "SpeechConfigParam": ".speechconfig",
     "StaticMediaProcessing": ".staticmediaprocessing",

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

@@ -19,6 +19,7 @@
 from __future__ import annotations
 from .. import BaseModel, UNSET_SENTINEL
 from .imageconfig import ImageConfig, ImageConfigParam
+from .speakerconfig import SpeakerConfig, SpeakerConfigParam
 from .speechconfig import SpeechConfig, SpeechConfigParam
 from .thinkinglevel import ThinkingLevel
 from .thinkingsummaries import ThinkingSummaries
@@ -42,6 +43,18 @@
 r"""The tool choice configuration."""
 
 
+SpeechConfigUnionParam = TypeAliasType(
+    "SpeechConfigUnionParam", Union[SpeakerConfigParam, List[SpeechConfigParam]]
+)
+r"""Optional. Speech and multi-speaker configuration."""
+
+
+SpeechConfigUnion = TypeAliasType(
+    "SpeechConfigUnion", Union[SpeakerConfig, List[SpeechConfig]]
+)
+r"""Optional. Speech and multi-speaker configuration."""
+
+
 class GenerationConfigParam(TypedDict):
     r"""Configuration parameters for model interactions."""
 
@@ -51,8 +64,6 @@ class GenerationConfigParam(TypedDict):
     r"""The maximum number of tokens to include in the response."""
     seed: NotRequired[int]
     r"""Seed used in decoding for reproducibility."""
-    speech_config: NotRequired[List[SpeechConfigParam]]
-    r"""Configuration for speech interaction."""
     stop_sequences: NotRequired[List[str]]
     r"""A list of character sequences that will stop output interaction."""
     thinking_level: NotRequired[ThinkingLevel]
@@ -63,6 +74,8 @@ class GenerationConfigParam(TypedDict):
     r"""Configuration for speech recognition (transcription)."""
     video_config: NotRequired[VideoConfigParam]
     r"""Configuration options for video generation."""
+    speech_config: NotRequired[SpeechConfigUnionParam]
+    r"""Optional. Speech and multi-speaker configuration."""
 
 
 class GenerationConfig(BaseModel):
@@ -82,9 +95,6 @@ class GenerationConfig(BaseModel):
     seed: Optional[int] = None
     r"""Seed used in decoding for reproducibility."""
 
-    speech_config: Optional[List[SpeechConfig]] = None
-    r"""Configuration for speech interaction."""
-
     stop_sequences: Optional[List[str]] = None
     r"""A list of character sequences that will stop output interaction."""
 
@@ -101,20 +111,23 @@ class GenerationConfig(BaseModel):
     video_config: Optional[VideoConfig] = None
     r"""Configuration options for video generation."""
 
+    speech_config: Optional[SpeechConfigUnion] = None
+    r"""Optional. Speech and multi-speaker configuration."""
+
     @model_serializer(mode="wrap")
     def serialize_model(self, handler):
         optional_fields = set(
             [
                 "image_config",
                 "max_output_tokens",
                 "seed",
-                "speech_config",
                 "stop_sequences",
                 "thinking_level",
                 "thinking_summaries",
                 "tool_choice",
                 "transcription_config",
                 "video_config",
+                "speech_config",
             ]
         )
         serialized = handler(self)

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

@@ -0,0 +1,54 @@
+# 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 BaseModel, UNSET_SENTINEL
+from .speechconfig import SpeechConfig, SpeechConfigParam
+from pydantic import model_serializer
+from typing import List, Optional
+from typing_extensions import NotRequired, TypedDict
+
+
+class SpeakerConfigParam(TypedDict):
+    r"""Configuration for multi-speaker and speech generation."""
+
+    speakers: NotRequired[List[SpeechConfigParam]]
+    r"""Individual speaker configurations."""
+
+
+class SpeakerConfig(BaseModel):
+    r"""Configuration for multi-speaker and speech generation."""
+
+    speakers: Optional[List[SpeechConfig]] = None
+    r"""Individual speaker configurations."""
+
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["speakers"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k, serialized.get(n))
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m