diff --git a/README.md b/README.md
index 89d5e8b..f7ea2aa 100644
--- a/README.md
+++ b/README.md
@@ -1,3 +1,107 @@
-# genreate
+# Generate
 
-文本生成，图像生成，语音生成
+> One API to Access World-Class Generative Models.
+
+## 简介
+
+Generate Package 允许用户通过统一的 api 访问跨平台的生成式模型，当前支持：
+
+* [OpenAI](https://platform.openai.com/docs/introduction)
+* [Azure](https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/chatgpt?tabs=python&pivots=programming-language-chat-completions)
+* [阿里云-百炼](https://bailian.console.aliyun.com/)
+* [百川智能](https://platform.baichuan-ai.com/docs/api)
+* [腾讯云-混元](https://cloud.tencent.com/document/product/1729)
+* [Minimax](https://api.minimax.chat/document/guides/chat)
+* [百度智能云](https://cloud.baidu.com/doc/WENXINWORKSHOP/s/clntwmv7t)
+* [智谱](https://open.bigmodel.cn/dev/api)
+* ...
+
+## Features
+
+* **多模态**，支持文本生成，图像生成以及语音生成
+* **跨平台**，支持 OpenAI，Azure，Minimax 在内的多家平台
+* **One API**，统一了不同平台的消息格式，推理参数，接口封装，返回解析
+* **异步和流式**，提供流式调用，非流式调用，同步调用，异步调用，适配不同的应用场景
+* **自带电池**，提供输入检查，参数检查，计费，*ChatEngine*, *CompletionEngine*, *function* 等功能
+* **高质量代码**，100% typehints，pylance strict, ruff lint & format,  test coverage > 85% ...
+
+## 基础使用
+
+### 安装
+
+```bash
+pip install generate-core
+```
+
+### 文本生成
+
+```python
+from generate import OpenAIChat
+
+model = OpenAIChat()
+model.generate('你好，GPT！', temperature=0, seed=2023)
+
+# ----- Output -----
+ChatCompletionOutput(
+    model_info=ModelInfo(task='chat_completion', type='openai', name='gpt-3.5-turbo-0613'),
+    cost=0.000343,
+    extra={'usage': {'prompt_tokens': 13, 'completion_tokens': 18, 'total_tokens': 31}},
+    messages=[
+        AssistantMessage(
+            content='你好！有什么我可以帮助你的吗？',
+            role='assistant',
+            name=None,
+            content_type='text'
+        )
+    ],
+    finish_reason='stop'
+)
+```
+
+### 图像生成
+
+```python
+from generate import OpenAIImageGeneration
+
+model = OpenAIImageGeneration()
+model.generate('black hole')
+
+# ----- Output -----
+ImageGenerationOutput(
+    model_info=ModelInfo(task='image_generation', type='openai', name='dall-e-3'),
+    cost=0.56,
+    extra={},
+    images=[
+        GeneratedImage(
+            url='https://oaidalleapiprodscus.blob.core.windows.net/...',
+            prompt='Visualize an astronomical illustration featuring a black hole at its core. The black hole 
+should be portrayed with strong gravitational lensing effect that distorts the light around it. Include a 
+surrounding accretion disk, glowing brightly with blue and white hues, streaked with shades of red and orange, 
+indicating heat and intense energy. The cosmos in the background should be filled with distant stars, galaxies, and
+nebulas, illuminating the vast, infinite space with specks of light.',
+            image_format='png',
+            content=b'<image bytes>'
+        )
+    ]
+)
+```
+
+### 语音生成
+
+```python
+from generate import MinimaxSpeech
+
+model = MinimaxSpeech()
+model.generate('你好，世界！')
+
+# ----- Output -----
+TextToSpeechOutput(
+    model_info=ModelInfo(task='text_to_speech', type='minimax', name='speech-01'),
+    cost=0.01,
+    extra={},
+    audio=b'<audio bytes>',
+    audio_format='mp3'
+)
+```
+
+‍
\ No newline at end of file
diff --git a/generate/__init__.py b/generate/__init__.py
index f7494c6..f7569ea 100644
--- a/generate/__init__.py
+++ b/generate/__init__.py
@@ -22,6 +22,14 @@
 from generate.chat_completion.function_call import function
 from generate.chat_engine import ChatEngine
 from generate.completion_engine import CompletionEngine
+from generate.image_generation import (
+    BaiduImageGeneration,
+    BaiduImageGenerationParameters,
+    OpenAIImageGeneration,
+    OpenAIImageGenerationParameters,
+    QianfanImageGeneration,
+    QianfanImageGenerationParameters,
+)
 from generate.text_to_speech import (
     MinimaxProSpeech,
     MinimaxProSpeechParameters,
@@ -68,6 +76,12 @@
     'MinimaxSpeechParameters',
     'MinimaxProSpeech',
     'MinimaxProSpeechParameters',
+    'OpenAIImageGeneration',
+    'OpenAIImageGenerationParameters',
+    'BaiduImageGeneration',
+    'BaiduImageGenerationParameters',
+    'QianfanImageGeneration',
+    'QianfanImageGenerationParameters',
     'function',
     'load_chat_model',
     'load_speech_model',
diff --git a/generate/chat_completion/__init__.py b/generate/chat_completion/__init__.py
index ec7be5a..1af5bd0 100644
--- a/generate/chat_completion/__init__.py
+++ b/generate/chat_completion/__init__.py
@@ -1,6 +1,6 @@
 from __future__ import annotations
 
-from typing import Type
+from typing import Any, Type
 
 from generate.chat_completion.base import ChatCompletionModel
 from generate.chat_completion.function_call import function, get_json_schema
@@ -29,7 +29,7 @@
 from generate.chat_completion.printer import MessagePrinter, SimpleMessagePrinter
 from generate.model import ModelParameters
 
-ChatModels: list[tuple[Type[ChatCompletionModel], Type[ModelParameters]]] = [
+ChatModels: list[tuple[Type[ChatCompletionModel[Any]], Type[ModelParameters]]] = [
     (AzureChat, OpenAIChatParameters),
     (OpenAIChat, OpenAIChatParameters),
     (MinimaxProChat, MinimaxProChatParameters),
@@ -42,7 +42,7 @@
     (BailianChat, BailianChatParameters),
 ]
 
-ChatModelRegistry: dict[str, tuple[Type[ChatCompletionModel], Type[ModelParameters]]] = {
+ChatModelRegistry: dict[str, tuple[Type[ChatCompletionModel[Any]], Type[ModelParameters]]] = {
     model_cls.model_type: (model_cls, parameter_cls) for model_cls, parameter_cls in ChatModels
 }
 
diff --git a/generate/chat_completion/function_call.py b/generate/chat_completion/function_call.py
index b1c6835..b73b720 100644
--- a/generate/chat_completion/function_call.py
+++ b/generate/chat_completion/function_call.py
@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 import json
-from typing import Callable, Generic, TypeVar
+from typing import Any, Callable, Dict, Generic, TypeVar, cast
 
 from docstring_parser import parse
 from pydantic import TypeAdapter, validate_call
@@ -20,7 +20,7 @@ class FunctionJsonSchema(TypedDict):
     description: NotRequired[str]
 
 
-def get_json_schema(function: Callable) -> FunctionJsonSchema:
+def get_json_schema(function: Callable[..., Any]) -> FunctionJsonSchema:
     function_name = function.__name__
     docstring = parse(text=function.__doc__ or '')
     parameters = TypeAdapter(function).json_schema()
@@ -71,7 +71,7 @@ def call_with_message(self, message: Message) -> T:
         raise ValueError(f'message is not a function call: {message}')
 
 
-def recusive_remove(dictionary: dict, remove_key: str) -> None:
+def recusive_remove(obj: Any, remove_key: str) -> None:
     """
     Recursively removes a key from a dictionary and all its nested dictionaries.
 
@@ -82,9 +82,10 @@ def recusive_remove(dictionary: dict, remove_key: str) -> None:
     Returns:
         None
     """
-    if isinstance(dictionary, dict):
-        for key in list(dictionary.keys()):
+    if isinstance(obj, dict):
+        obj = cast(Dict[str, Any], obj)
+        for key in list(obj.keys()):
             if key == remove_key:
-                del dictionary[key]
+                del obj[key]
             else:
-                recusive_remove(dictionary[key], remove_key)
+                recusive_remove(obj[key], remove_key)
diff --git a/generate/chat_completion/message/core.py b/generate/chat_completion/message/core.py
index 2c3302a..db163bb 100644
--- a/generate/chat_completion/message/core.py
+++ b/generate/chat_completion/message/core.py
@@ -7,8 +7,6 @@
 
 
 class Message(BaseModel):
-    role: str
-    name: Optional[str] = None
     content: Any
 
 
@@ -19,6 +17,7 @@ class SystemMessage(Message):
 
 class UserMessage(Message):
     role: Literal['user'] = 'user'
+    name: Optional[str] = None
     content_type: Literal['text'] = 'text'
     content: str
 
@@ -43,6 +42,7 @@ class ImageUrlPart(BaseModel):
 
 class UserMultiPartMessage(Message):
     role: Literal['user'] = 'user'
+    name: Optional[str] = None
     content_type: Literal['multi_part'] = 'multi_part'
     content: List[UserPartTypes]
 
@@ -62,6 +62,7 @@ class ToolMessage(Message):
 
 class AssistantMessage(Message):
     role: Literal['assistant'] = 'assistant'
+    name: Optional[str] = None
     content_type: Literal['text'] = 'text'
     content: str
 
@@ -74,6 +75,7 @@ class FunctionCall(BaseModel):
 
 class FunctionCallMessage(Message):
     role: Literal['assistant'] = 'assistant'
+    name: Optional[str] = None
     content_type: Literal['function_call'] = 'function_call'
     content: FunctionCall
 
@@ -86,6 +88,7 @@ class ToolCall(BaseModel):
 
 class ToolCallsMessage(Message):
     role: Literal['assistant'] = 'assistant'
+    name: Optional[str] = None
     content_type: Literal['tool_calls'] = 'tool_calls'
     content: List[ToolCall]
 
diff --git a/generate/chat_completion/message/utils.py b/generate/chat_completion/message/utils.py
index deea6ed..8de501c 100644
--- a/generate/chat_completion/message/utils.py
+++ b/generate/chat_completion/message/utils.py
@@ -55,15 +55,11 @@ def infer_assistant_message_content_type(message_content: Any) -> Literal['text'
         return 'text'
     if isinstance(obj, FunctionCall):
         return 'function_call'
-    if isinstance(obj, list):
-        return 'tool_calls'
-    raise ValueError(f'Unknown content type: {obj}')
+    return 'tool_calls'
 
 
 def infer_user_message_content_type(message_content: Any) -> Literal['text', 'multi_part']:
     obj = user_content_validator.validate_python(message_content)
     if isinstance(obj, str):
         return 'text'
-    if isinstance(obj, list):
-        return 'multi_part'
-    raise ValueError(f'Unknown content type: {obj}')
+    return 'multi_part'
diff --git a/generate/chat_completion/models/azure.py b/generate/chat_completion/models/azure.py
index 23f3fad..15aaff1 100644
--- a/generate/chat_completion/models/azure.py
+++ b/generate/chat_completion/models/azure.py
@@ -73,7 +73,7 @@ def _stream_completion(self, messages: Messages, parameters: OpenAIChatParameter
         raise NotImplementedError('Azure does not support streaming')
 
     @override
-    async def _async_stream_completion(
+    def _async_stream_completion(
         self, messages: Messages, parameters: OpenAIChatParameters
     ) -> AsyncIterator[ChatCompletionStreamOutput]:
         raise NotImplementedError('Azure does not support streaming')
diff --git a/generate/chat_completion/models/baichuan.py b/generate/chat_completion/models/baichuan.py
index b3c849f..879bc61 100644
--- a/generate/chat_completion/models/baichuan.py
+++ b/generate/chat_completion/models/baichuan.py
@@ -77,7 +77,7 @@ def __init__(
 
     def _get_request_parameters(self, messages: Messages, parameters: BaichuanChatParameters) -> HttpxPostKwargs:
         baichuan_messages: list[BaichuanMessage] = [convert_to_baichuan_message(message) for message in messages]
-        data = {
+        data: dict[str, Any] = {
             'model': self.model,
             'messages': baichuan_messages,
         }
diff --git a/generate/chat_completion/models/bailian.py b/generate/chat_completion/models/bailian.py
index 380dbef..9912a32 100644
--- a/generate/chat_completion/models/bailian.py
+++ b/generate/chat_completion/models/bailian.py
@@ -43,7 +43,7 @@ def generate_default_request_id() -> str:
 
 
 def convert_to_bailian_chat_qa_pair(messages: Messages) -> list[BailianChatQAPair]:
-    pairs = []
+    pairs: list[BailianChatQAPair] = []
     for user_message, assistant_message in zip(messages[::2], messages[1::2]):
         if not isinstance(user_message, UserMessage):
             raise MessageTypeError(user_message, allowed_message_type=(UserMessage,))
diff --git a/generate/chat_completion/models/minimax_pro.py b/generate/chat_completion/models/minimax_pro.py
index 43bc18c..2b81462 100644
--- a/generate/chat_completion/models/minimax_pro.py
+++ b/generate/chat_completion/models/minimax_pro.py
@@ -146,8 +146,6 @@ def convert_to_minimax_pro_message(
         }
 
     if isinstance(message, FunctionMessage):
-        if message.name is None:
-            raise MessageValueError(message, 'function name is required')
         return {
             'sender_type': 'FUNCTION',
             'sender_name': message.name,
diff --git a/generate/chat_completion/models/openai.py b/generate/chat_completion/models/openai.py
index 053d635..fdb5a3d 100644
--- a/generate/chat_completion/models/openai.py
+++ b/generate/chat_completion/models/openai.py
@@ -14,11 +14,9 @@
     FunctionCall,
     FunctionCallMessage,
     FunctionMessage,
-    ImageUrlPart,
     Message,
     Messages,
     MessageTypeError,
-    MessageValueError,
     SystemMessage,
     TextPart,
     ToolCall,
@@ -65,7 +63,7 @@ class OpenAIToolCall(TypedDict):
 
 class OpenAIMessage(TypedDict):
     role: str
-    content: Union[str, None, List[Dict]]
+    content: Union[str, None, List[Dict[str, Any]]]
     name: NotRequired[str]
     function_call: NotRequired[OpenAIFunctionCall]
     tool_call_id: NotRequired[str]
@@ -101,12 +99,12 @@ def _to_text_message_dict(role: str, message: Message) -> OpenAIMessage:
 
 
 def _to_user_multipart_message_dict(message: UserMultiPartMessage) -> OpenAIMessage:
-    content = []
+    content: list[dict[str, Any]] = []
     for part in message.content:
         if isinstance(part, TextPart):
             content.append({'type': 'text', 'text': part.text})
-        elif isinstance(part, ImageUrlPart):
-            image_url_part_dict = {
+        else:
+            image_url_part_dict: dict[str, Any] = {
                 'type': 'image_url',
                 'image_url': {
                     'url': part.image_url.url,
@@ -115,9 +113,6 @@ def _to_user_multipart_message_dict(message: UserMultiPartMessage) -> OpenAIMess
             if part.image_url.detail:
                 image_url_part_dict['image_url']['detail'] = part.image_url.detail
             content.append(image_url_part_dict)
-        else:
-            raise MessageValueError(message, f'OpenAI does not support {type(part)} ')
-
     return {
         'role': 'user',
         'content': content,
diff --git a/generate/chat_engine.py b/generate/chat_engine.py
index b29729b..4f86c3a 100644
--- a/generate/chat_engine.py
+++ b/generate/chat_engine.py
@@ -25,7 +25,7 @@
 
 
 class ChatEngineKwargs(TypedDict, total=False):
-    functions: Union[List[function], dict[str, Callable], None]
+    functions: Union[List[function[Any, Any]], dict[str, Callable[..., Any]], None]
     call_raise_error: bool
     max_calls_per_turn: int
 
@@ -51,7 +51,7 @@ class ChatEngine(Generic[P]):
     def __init__(
         self,
         chat_model: ChatCompletionModel[P],
-        functions: Union[List[function], dict[str, Callable], None] = None,
+        functions: Union[List[function[Any, Any]], dict[str, Callable[..., Any]], None] = None,
         call_raise_error: bool = False,
         max_calls_per_turn: int = 5,
         stream: bool | Literal['auto'] = 'auto',
@@ -60,10 +60,8 @@ def __init__(
         self._chat_model = chat_model
 
         if isinstance(functions, list):
-            self._function_map = {}
+            self._function_map: dict[str, Callable[..., Any]] = {}
             for i in functions:
-                if not isinstance(i, function):
-                    raise TypeError(f'Invalid function type: {type(i)}')
                 self._function_map[i.json_schema['name']] = i
         elif isinstance(functions, dict):
             self._function_map = functions
@@ -127,7 +125,7 @@ def handle_model_output(self, model_output: ChatCompletionOutput, **override_par
         self.history.extend(model_output.messages)
         if self.printer:
             for message in model_output.messages:
-                if self.stream and message.role == 'assistant':
+                if self.stream and isinstance(message, AssistantMessage):
                     continue
                 self.printer.print_message(message)
 
diff --git a/generate/http.py b/generate/http.py
index 4dd8022..e55a06c 100644
--- a/generate/http.py
+++ b/generate/http.py
@@ -48,7 +48,7 @@ class UnexpectedResponseError(Exception):
         response (dict): The response from the server.
     """
 
-    def __init__(self, response: dict, *args: Any) -> None:
+    def __init__(self, response: dict[str, Any], *args: Any) -> None:
         super().__init__(response, *args)
 
 
diff --git a/generate/image_generation/__init__.py b/generate/image_generation/__init__.py
index c9bb9ac..c1041b9 100644
--- a/generate/image_generation/__init__.py
+++ b/generate/image_generation/__init__.py
@@ -1,6 +1,6 @@
 from __future__ import annotations
 
-from typing import Type
+from typing import Any, Type, TypeVar
 
 from generate.image_generation.base import GeneratedImage, ImageGenerationModel, ImageGenerationOutput
 from generate.image_generation.models import (
@@ -8,14 +8,18 @@
     BaiduImageGenerationParameters,
     OpenAIImageGeneration,
     OpenAIImageGenerationParameters,
+    QianfanImageGeneration,
+    QianfanImageGenerationParameters,
 )
 from generate.model import ModelParameters
 
-ImageGenerationModels: list[tuple[Type[ImageGenerationModel], Type[ModelParameters]]] = [
+P = TypeVar('P', bound=ModelParameters)
+
+ImageGenerationModels: list[tuple[Type[ImageGenerationModel[Any]], Type[ModelParameters]]] = [
     (OpenAIImageGeneration, OpenAIImageGenerationParameters),
 ]
 
-ImageGenerationModelRegistry: dict[str, tuple[Type[ImageGenerationModel], Type[ModelParameters]]] = {
+ImageGenerationModelRegistry: dict[str, tuple[Type[ImageGenerationModel[Any]], Type[ModelParameters]]] = {
     model_cls.model_type: (model_cls, parameter_cls) for model_cls, parameter_cls in ImageGenerationModels
 }
 
@@ -27,4 +31,6 @@
     'OpenAIImageGenerationParameters',
     'BaiduImageGeneration',
     'BaiduImageGenerationParameters',
+    'QianfanImageGeneration',
+    'QianfanImageGenerationParameters',
 ]
diff --git a/generate/image_generation/models/__init__.py b/generate/image_generation/models/__init__.py
index 9538121..37c6e2b 100644
--- a/generate/image_generation/models/__init__.py
+++ b/generate/image_generation/models/__init__.py
@@ -1,9 +1,12 @@
 from generate.image_generation.models.baidu import BaiduImageGeneration, BaiduImageGenerationParameters
 from generate.image_generation.models.openai import OpenAIImageGeneration, OpenAIImageGenerationParameters
+from generate.image_generation.models.qianfan import QianfanImageGeneration, QianfanImageGenerationParameters
 
 __all__ = [
     'OpenAIImageGeneration',
     'OpenAIImageGenerationParameters',
     'BaiduImageGeneration',
     'BaiduImageGenerationParameters',
+    'QianfanImageGeneration',
+    'QianfanImageGenerationParameters',
 ]
diff --git a/generate/image_generation/models/baidu.py b/generate/image_generation/models/baidu.py
index 9881f50..18afc4b 100644
--- a/generate/image_generation/models/baidu.py
+++ b/generate/image_generation/models/baidu.py
@@ -33,7 +33,7 @@ class BaiduImageGenerationParameters(ModelParameters):
     change_degree: Optional[Annotated[int, Field(ge=1, le=10)]] = None
 
     def custom_model_dump(self) -> dict[str, Any]:
-        output_data = {}
+        output_data: dict[str, Any] = {}
         width, height = self.size.split('x')
         output_data['width'] = int(width)
         output_data['height'] = int(height)
@@ -146,7 +146,7 @@ def _parse_task_info(self, task_info: dict[str, Any]) -> list[str] | None:
         if task_status == 'FAILED':
             raise UnexpectedResponseError(task_info, 'Task failed')
         if task_status == 'SUCCESS':
-            image_urls = []
+            image_urls: list[str] = []
             for sub_result in task_info['data']['sub_task_result_list']:
                 image_url = sub_result['final_image_list'][0]['img_url']
                 image_urls.append(image_url)
@@ -155,7 +155,7 @@ def _parse_task_info(self, task_info: dict[str, Any]) -> list[str] | None:
 
     def _get_image(self, task_id: str) -> list[str]:
         start_time = time.time()
-        task_info = {}
+        task_info: dict[str, Any] = {}
         while (time.time() - start_time) < self.task_timeout:
             response = self.http_client.post(self._get_image_request_parameters(task_id))
             task_info = response.json()
@@ -167,7 +167,7 @@ def _get_image(self, task_id: str) -> list[str]:
 
     async def _async_get_image(self, task_id: str) -> list[str]:
         start_time = time.time()
-        task_info = {}
+        task_info: dict[str, Any] = {}
         while (time.time() - start_time) < self.task_timeout:
             response = await self.http_client.async_post(self._get_image_request_parameters(task_id))
             task_info = response.json()
diff --git a/generate/image_generation/models/openai.py b/generate/image_generation/models/openai.py
index 49f4c66..f570ef0 100644
--- a/generate/image_generation/models/openai.py
+++ b/generate/image_generation/models/openai.py
@@ -118,7 +118,7 @@ def _construct_model_output(
         self, prompt: str, parameters: OpenAIImageGenerationParameters, response: Response
     ) -> ImageGenerationOutput:
         response_data = response.json()
-        generated_images = []
+        generated_images: list[GeneratedImage] = []
         for image_data in response_data['data']:
             image_prompt = image_data.get('revised_prompt') or prompt
             url = image_data.get('url')
@@ -140,6 +140,7 @@ def _construct_model_output(
         return ImageGenerationOutput(
             model_info=self.model_info,
             images=generated_images,
+            cost=self.calculate_cost(parameters),
         )
 
     def calculate_cost(self, parameters: OpenAIImageGenerationParameters) -> float:
diff --git a/generate/image_generation/models/qianfan.py b/generate/image_generation/models/qianfan.py
index f44d578..626bf7c 100644
--- a/generate/image_generation/models/qianfan.py
+++ b/generate/image_generation/models/qianfan.py
@@ -31,7 +31,7 @@ class QianfanImageGenerationParameters(ModelParameters):
 
 
 class QianfanImageGeneration(ImageGenerationModel[QianfanImageGenerationParameters], QianfanTokenMixin):
-    model_type = 'baidu'
+    model_type = 'qianfan'
 
     def __init__(
         self,
diff --git a/generate/model.py b/generate/model.py
index 7629e15..2704597 100644
--- a/generate/model.py
+++ b/generate/model.py
@@ -30,7 +30,7 @@ class ModelOutput(BaseModel):
     extra: Dict[str, Any] = {}
 
     @model_serializer(mode='wrap')
-    def ser_model(self, handler) -> Any:  # noqa: ANN001
+    def ser_model(self, handler: Any) -> Any:  # noqa: ANN001
         output = handler(self)
         output.pop('debug')
         return output
diff --git a/generate/platforms/baidu.py b/generate/platforms/baidu.py
index bea9a2c..176cd7f 100644
--- a/generate/platforms/baidu.py
+++ b/generate/platforms/baidu.py
@@ -8,7 +8,7 @@
 
 
 class QianfanSettings(BaseSettings):
-    model_config = SettingsConfigDict(extra='ignore', env_prefix='wenxin_', env_file='.env')
+    model_config = SettingsConfigDict(extra='ignore', env_prefix='qianfan_', env_file='.env')
 
     api_key: SecretStr
     secret_key: SecretStr
diff --git a/generate/test.py b/generate/test.py
index 5d85fcb..315c70a 100644
--- a/generate/test.py
+++ b/generate/test.py
@@ -14,16 +14,16 @@ def get_pytest_params(
     model_registry: Mapping[str, tuple[Any, Type[ModelParameters]]],
     types: Sequence[param_type] | param_type,
     exclude: Sequence[str] | None = None,
-) -> list:
+) -> list[Any]:
     exclude = exclude or []
     if isinstance(types, str):
         types = [types]
 
-    pytest_params: list = []
+    pytest_params: list[Any] = []
     for model_name, (model_cls, paramter_cls) in model_registry.items():
         if model_name in exclude:
             continue
-        values = []
+        values: list[Any] = []
         for t in types:
             if t == 'model':
                 values.append(model_cls(parameters=paramter_cls()))
diff --git a/generate/text_to_speech/__init__.py b/generate/text_to_speech/__init__.py
index 6344c36..97925c0 100644
--- a/generate/text_to_speech/__init__.py
+++ b/generate/text_to_speech/__init__.py
@@ -1,6 +1,6 @@
 from __future__ import annotations
 
-from typing import Type
+from typing import Any, Type
 
 from generate.model import ModelParameters
 from generate.text_to_speech.base import TextToSpeechModel, TextToSpeechOutput
@@ -13,13 +13,13 @@
     OpenAISpeechParameters,
 )
 
-SpeechModels: list[tuple[Type[TextToSpeechModel], Type[ModelParameters]]] = [
+SpeechModels: list[tuple[Type[TextToSpeechModel[Any]], Type[ModelParameters]]] = [
     (MinimaxSpeech, MinimaxSpeechParameters),
     (MinimaxProSpeech, MinimaxProSpeechParameters),
     (OpenAISpeech, OpenAISpeechParameters),
 ]
 
-SpeechModelRegistry: dict[str, tuple[Type[TextToSpeechModel], Type[ModelParameters]]] = {
+SpeechModelRegistry: dict[str, tuple[Type[TextToSpeechModel[Any]], Type[ModelParameters]]] = {
     model_cls.model_type: (model_cls, parameter_cls) for model_cls, parameter_cls in SpeechModels
 }
 
diff --git a/generate/text_to_speech/base.py b/generate/text_to_speech/base.py
index 6b5dd80..ba2b24e 100644
--- a/generate/text_to_speech/base.py
+++ b/generate/text_to_speech/base.py
@@ -29,12 +29,12 @@ def _text_to_speech(self, text: str, parameters: P) -> TextToSpeechOutput:
     async def _async_text_to_speech(self, text: str, parameters: P) -> TextToSpeechOutput:
         ...
 
-    def generate(self, text: str, **override_parameters: Any) -> TextToSpeechOutput:
+    def generate(self, prompt: str, **override_parameters: Any) -> TextToSpeechOutput:
         parameters = self._merge_parameters(**override_parameters)
-        logger.debug(f'{text=}, {parameters=}')
-        return self._text_to_speech(text, parameters)
+        logger.debug(f'{prompt=}, {parameters=}')
+        return self._text_to_speech(prompt, parameters)
 
-    async def async_generate(self, text: str, **override_parameters: Any) -> TextToSpeechOutput:
+    async def async_generate(self, prompt: str, **override_parameters: Any) -> TextToSpeechOutput:
         parameters = self._merge_parameters(**override_parameters)
-        logger.debug(f'{text=}, {parameters=}')
-        return await self._async_text_to_speech(text, parameters)
+        logger.debug(f'{prompt=}, {parameters=}')
+        return await self._async_text_to_speech(prompt, parameters)
diff --git a/generate/text_to_speech/models/minimax.py b/generate/text_to_speech/models/minimax.py
index f8bf66d..2f31a3b 100644
--- a/generate/text_to_speech/models/minimax.py
+++ b/generate/text_to_speech/models/minimax.py
@@ -17,7 +17,7 @@ class TimeberWeight(TypedDict):
 
 
 class MinimaxSpeechParameters(ModelParameters):
-    voice_id: Optional[str] = None
+    voice: Optional[str] = Field(default=None, alias='voice_id')
     speed: Annotated[Optional[float], Field(ge=0.5, le=2.0)] = None
     vol: Annotated[Optional[float], Field(gt=0, le=10)] = None
     pitch: Annotated[Optional[float], Field(ge=-12, le=12)] = None
@@ -25,8 +25,8 @@ class MinimaxSpeechParameters(ModelParameters):
 
     @model_validator(mode='after')
     def voice_exists(self) -> Self:
-        if self.voice_id is None and self.timber_weights is None:
-            self.voice_id = 'male-qn-qingse'
+        if self.voice is None and self.timber_weights is None:
+            self.voice = 'male-qn-qingse'
         return self
 
 
diff --git a/generate/text_to_speech/models/openai.py b/generate/text_to_speech/models/openai.py
index cf576a7..d6ea43b 100644
--- a/generate/text_to_speech/models/openai.py
+++ b/generate/text_to_speech/models/openai.py
@@ -12,7 +12,7 @@
 
 
 class OpenAISpeechParameters(ModelParameters):
-    voice_id: Annotated[str, Field(serialization_alias='voice')] = 'alloy'
+    voice: str = 'alloy'
     response_format: Optional[Literal['mp3', 'aac', 'opus', 'flac']] = None
     speed: Annotated[Optional[float], Field(ge=0.25, le=4.0)] = None
 
diff --git a/generate/utils.py b/generate/utils.py
index 3705955..cfee906 100644
--- a/generate/utils.py
+++ b/generate/utils.py
@@ -8,7 +8,7 @@
 from generate.text_to_speech import SpeechModelRegistry, TextToSpeechModel, TextToSpeechOutput
 
 
-def load_chat_model(model_id: str, **kwargs: Any) -> ChatCompletionModel:
+def load_chat_model(model_id: str, **kwargs: Any) -> ChatCompletionModel[Any]:
     if '/' not in model_id:
         model_type = model_id
         return ChatModelRegistry[model_type][0](**kwargs)  # type: ignore
@@ -17,7 +17,7 @@ def load_chat_model(model_id: str, **kwargs: Any) -> ChatCompletionModel:
     return model_cls.from_name(name, **kwargs)
 
 
-def load_speech_model(speech_model_id: str, **kwargs: Any) -> TextToSpeechModel:
+def load_speech_model(speech_model_id: str, **kwargs: Any) -> TextToSpeechModel[Any]:
     if '/' not in speech_model_id:
         model_type = speech_model_id
         return SpeechModelRegistry[model_type][0](**kwargs)  # type: ignore
@@ -26,7 +26,7 @@ def load_speech_model(speech_model_id: str, **kwargs: Any) -> TextToSpeechModel:
     return model_cls.from_name(name, **kwargs)
 
 
-def load_image_generation_model(model_id: str, **kwargs: Any) -> ImageGenerationModel:
+def load_image_generation_model(model_id: str, **kwargs: Any) -> ImageGenerationModel[Any]:
     if '/' not in model_id:
         model_type = model_id
         return ImageGenerationModelRegistry[model_type][0](**kwargs)  # type: ignore
diff --git a/poetry.lock b/poetry.lock
index 8aa54a7..49378cf 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -538,6 +538,70 @@ files = [
     {file = "colorama-0.4.6.tar.gz", hash = "sha256:08695f5cb7ed6e0531a20572697297273c47b8cae5a63ffc6d6ed5c201be6e44"},
 ]
 
+[[package]]
+name = "coverage"
+version = "7.3.2"
+description = "Code coverage measurement for Python"
+optional = false
+python-versions = ">=3.8"
+files = [
+    {file = "coverage-7.3.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:d872145f3a3231a5f20fd48500274d7df222e291d90baa2026cc5152b7ce86bf"},
+    {file = "coverage-7.3.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:310b3bb9c91ea66d59c53fa4989f57d2436e08f18fb2f421a1b0b6b8cc7fffda"},
+    {file = "coverage-7.3.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f47d39359e2c3779c5331fc740cf4bce6d9d680a7b4b4ead97056a0ae07cb49a"},
+    {file = "coverage-7.3.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:aa72dbaf2c2068404b9870d93436e6d23addd8bbe9295f49cbca83f6e278179c"},
+    {file = "coverage-7.3.2-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:beaa5c1b4777f03fc63dfd2a6bd820f73f036bfb10e925fce067b00a340d0f3f"},
+    {file = "coverage-7.3.2-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:dbc1b46b92186cc8074fee9d9fbb97a9dd06c6cbbef391c2f59d80eabdf0faa6"},
+    {file = "coverage-7.3.2-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:315a989e861031334d7bee1f9113c8770472db2ac484e5b8c3173428360a9148"},
+    {file = "coverage-7.3.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:d1bc430677773397f64a5c88cb522ea43175ff16f8bfcc89d467d974cb2274f9"},
+    {file = "coverage-7.3.2-cp310-cp310-win32.whl", hash = "sha256:a889ae02f43aa45032afe364c8ae84ad3c54828c2faa44f3bfcafecb5c96b02f"},
+    {file = "coverage-7.3.2-cp310-cp310-win_amd64.whl", hash = "sha256:c0ba320de3fb8c6ec16e0be17ee1d3d69adcda99406c43c0409cb5c41788a611"},
+    {file = "coverage-7.3.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:ac8c802fa29843a72d32ec56d0ca792ad15a302b28ca6203389afe21f8fa062c"},
+    {file = "coverage-7.3.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:89a937174104339e3a3ffcf9f446c00e3a806c28b1841c63edb2b369310fd074"},
+    {file = "coverage-7.3.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e267e9e2b574a176ddb983399dec325a80dbe161f1a32715c780b5d14b5f583a"},
+    {file = "coverage-7.3.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2443cbda35df0d35dcfb9bf8f3c02c57c1d6111169e3c85fc1fcc05e0c9f39a3"},
+    {file = "coverage-7.3.2-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4175e10cc8dda0265653e8714b3174430b07c1dca8957f4966cbd6c2b1b8065a"},
+    {file = "coverage-7.3.2-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:0cbf38419fb1a347aaf63481c00f0bdc86889d9fbf3f25109cf96c26b403fda1"},
+    {file = "coverage-7.3.2-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:5c913b556a116b8d5f6ef834038ba983834d887d82187c8f73dec21049abd65c"},
+    {file = "coverage-7.3.2-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:1981f785239e4e39e6444c63a98da3a1db8e971cb9ceb50a945ba6296b43f312"},
+    {file = "coverage-7.3.2-cp311-cp311-win32.whl", hash = "sha256:43668cabd5ca8258f5954f27a3aaf78757e6acf13c17604d89648ecc0cc66640"},
+    {file = "coverage-7.3.2-cp311-cp311-win_amd64.whl", hash = "sha256:e10c39c0452bf6e694511c901426d6b5ac005acc0f78ff265dbe36bf81f808a2"},
+    {file = "coverage-7.3.2-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:4cbae1051ab791debecc4a5dcc4a1ff45fc27b91b9aee165c8a27514dd160836"},
+    {file = "coverage-7.3.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:12d15ab5833a997716d76f2ac1e4b4d536814fc213c85ca72756c19e5a6b3d63"},
+    {file = "coverage-7.3.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3c7bba973ebee5e56fe9251300c00f1579652587a9f4a5ed8404b15a0471f216"},
+    {file = "coverage-7.3.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:fe494faa90ce6381770746077243231e0b83ff3f17069d748f645617cefe19d4"},
+    {file = "coverage-7.3.2-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f6e9589bd04d0461a417562649522575d8752904d35c12907d8c9dfeba588faf"},
+    {file = "coverage-7.3.2-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:d51ac2a26f71da1b57f2dc81d0e108b6ab177e7d30e774db90675467c847bbdf"},
+    {file = "coverage-7.3.2-cp312-cp312-musllinux_1_1_i686.whl", hash = "sha256:99b89d9f76070237975b315b3d5f4d6956ae354a4c92ac2388a5695516e47c84"},
+    {file = "coverage-7.3.2-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:fa28e909776dc69efb6ed975a63691bc8172b64ff357e663a1bb06ff3c9b589a"},
+    {file = "coverage-7.3.2-cp312-cp312-win32.whl", hash = "sha256:289fe43bf45a575e3ab10b26d7b6f2ddb9ee2dba447499f5401cfb5ecb8196bb"},
+    {file = "coverage-7.3.2-cp312-cp312-win_amd64.whl", hash = "sha256:7dbc3ed60e8659bc59b6b304b43ff9c3ed858da2839c78b804973f613d3e92ed"},
+    {file = "coverage-7.3.2-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:f94b734214ea6a36fe16e96a70d941af80ff3bfd716c141300d95ebc85339738"},
+    {file = "coverage-7.3.2-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:af3d828d2c1cbae52d34bdbb22fcd94d1ce715d95f1a012354a75e5913f1bda2"},
+    {file = "coverage-7.3.2-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:630b13e3036e13c7adc480ca42fa7afc2a5d938081d28e20903cf7fd687872e2"},
+    {file = "coverage-7.3.2-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c9eacf273e885b02a0273bb3a2170f30e2d53a6d53b72dbe02d6701b5296101c"},
+    {file = "coverage-7.3.2-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d8f17966e861ff97305e0801134e69db33b143bbfb36436efb9cfff6ec7b2fd9"},
+    {file = "coverage-7.3.2-cp38-cp38-musllinux_1_1_aarch64.whl", hash = "sha256:b4275802d16882cf9c8b3d057a0839acb07ee9379fa2749eca54efbce1535b82"},
+    {file = "coverage-7.3.2-cp38-cp38-musllinux_1_1_i686.whl", hash = "sha256:72c0cfa5250f483181e677ebc97133ea1ab3eb68645e494775deb6a7f6f83901"},
+    {file = "coverage-7.3.2-cp38-cp38-musllinux_1_1_x86_64.whl", hash = "sha256:cb536f0dcd14149425996821a168f6e269d7dcd2c273a8bff8201e79f5104e76"},
+    {file = "coverage-7.3.2-cp38-cp38-win32.whl", hash = "sha256:307adb8bd3abe389a471e649038a71b4eb13bfd6b7dd9a129fa856f5c695cf92"},
+    {file = "coverage-7.3.2-cp38-cp38-win_amd64.whl", hash = "sha256:88ed2c30a49ea81ea3b7f172e0269c182a44c236eb394718f976239892c0a27a"},
+    {file = "coverage-7.3.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:b631c92dfe601adf8f5ebc7fc13ced6bb6e9609b19d9a8cd59fa47c4186ad1ce"},
+    {file = "coverage-7.3.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:d3d9df4051c4a7d13036524b66ecf7a7537d14c18a384043f30a303b146164e9"},
+    {file = "coverage-7.3.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5f7363d3b6a1119ef05015959ca24a9afc0ea8a02c687fe7e2d557705375c01f"},
+    {file = "coverage-7.3.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2f11cc3c967a09d3695d2a6f03fb3e6236622b93be7a4b5dc09166a861be6d25"},
+    {file = "coverage-7.3.2-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:149de1d2401ae4655c436a3dced6dd153f4c3309f599c3d4bd97ab172eaf02d9"},
+    {file = "coverage-7.3.2-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:3a4006916aa6fee7cd38db3bfc95aa9c54ebb4ffbfc47c677c8bba949ceba0a6"},
+    {file = "coverage-7.3.2-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:9028a3871280110d6e1aa2df1afd5ef003bab5fb1ef421d6dc748ae1c8ef2ebc"},
+    {file = "coverage-7.3.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:9f805d62aec8eb92bab5b61c0f07329275b6f41c97d80e847b03eb894f38d083"},
+    {file = "coverage-7.3.2-cp39-cp39-win32.whl", hash = "sha256:d1c88ec1a7ff4ebca0219f5b1ef863451d828cccf889c173e1253aa84b1e07ce"},
+    {file = "coverage-7.3.2-cp39-cp39-win_amd64.whl", hash = "sha256:b4767da59464bb593c07afceaddea61b154136300881844768037fd5e859353f"},
+    {file = "coverage-7.3.2-pp38.pp39.pp310-none-any.whl", hash = "sha256:ae97af89f0fbf373400970c0a21eef5aa941ffeed90aee43650b81f7d7f47637"},
+    {file = "coverage-7.3.2.tar.gz", hash = "sha256:be32ad29341b0170e795ca590e1c07e81fc061cb5b10c74ce7203491484404ef"},
+]
+
+[package.extras]
+toml = ["tomli"]
+
 [[package]]
 name = "cryptography"
 version = "41.0.5"
@@ -1354,4 +1418,4 @@ multidict = ">=4.0"
 [metadata]
 lock-version = "2.0"
 python-versions = "^3.8"
-content-hash = "d4ded45bfcab127ad80aecf3e5c88064988e539a6418c71250d55417cf3cdd03"
+content-hash = "ff092b5a43fe8982f5684ffaa8f3944092e21230c836827786f6e022f42f5b9a"
diff --git a/pyproject.toml b/pyproject.toml
index 26a6b32..2f3900b 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -69,6 +69,7 @@ reportMissingTypeStubs=false
 pytest = "^7.3.1"
 pyright = "^1.1.310"
 ruff = "^0.1.4"
+coverage = "^7.3.2"
 
 [tool.poetry.group.bailian.dependencies]
 broadscope-bailian = "^1.1.8"
diff --git a/tests/test_chat_completion_model.py b/tests/test_chat_completion_model.py
index a360495..013eb28 100644
--- a/tests/test_chat_completion_model.py
+++ b/tests/test_chat_completion_model.py
@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 import asyncio
-from typing import Type
+from typing import Any, Type
 
 import pytest
 
@@ -31,7 +31,9 @@ def test_model_type_is_unique() -> None:
         {'temperature': 0.5, 'top_p': 0.85, 'max_tokens': 20},
     ],
 )
-def test_http_chat_model(model_cls: Type[ChatCompletionModel], parameter_cls: Type[ModelParameters], parameters: dict) -> None:
+def test_http_chat_model(
+    model_cls: Type[ChatCompletionModel[Any]], parameter_cls: Type[ModelParameters], parameters: dict[str, Any]
+) -> None:
     model = model_cls(parameters=parameter_cls())
     prompt = '这是测试，只回复你好'
     sync_output = model.generate(prompt, **parameters)
@@ -45,7 +47,7 @@ def test_http_chat_model(model_cls: Type[ChatCompletionModel], parameter_cls: Ty
     'chat_completion_model',
     get_pytest_params('test_stream_chat_completion', ChatModelRegistry, types='model', exclude=['azure']),
 )
-def test_http_stream_chat_model(chat_completion_model: ChatCompletionModel) -> None:
+def test_http_stream_chat_model(chat_completion_model: ChatCompletionModel[Any]) -> None:
     prompt = '这是测试，只回复你好'
     sync_output = list(chat_completion_model.stream_generate(prompt))[-1]
     async_output = asyncio.run(async_stream_helper(chat_completion_model, prompt))
@@ -55,7 +57,7 @@ def test_http_stream_chat_model(chat_completion_model: ChatCompletionModel) -> N
     assert async_output.reply != ''
 
 
-async def async_stream_helper(model: ChatCompletionModel, prompt: Prompt) -> ChatCompletionStreamOutput:
+async def async_stream_helper(model: ChatCompletionModel[Any], prompt: Prompt) -> ChatCompletionStreamOutput:
     async for output in model.async_stream_generate(prompt):
         if output.stream.control == 'finish':
             return output
@@ -69,7 +71,7 @@ async def async_stream_helper(model: ChatCompletionModel, prompt: Prompt) -> Cha
     ),
 )
 def test_init_chat_parameters(
-    model_cls: Type[ChatCompletionModel], parameters: ModelParameters, temperature: float = 0.8
+    model_cls: Type[ChatCompletionModel[Any]], parameters: ModelParameters, temperature: float = 0.8
 ) -> None:
     parameters.temperature = temperature
 
diff --git a/tests/test_completion.py b/tests/test_completion.py
index c5cbed9..759a32e 100644
--- a/tests/test_completion.py
+++ b/tests/test_completion.py
@@ -78,7 +78,7 @@ def test_sync_completion() -> None:
     assert len(results) == len(prompts)
 
 
-async def async_helper(client: CompletionEngine, prompts: Prompts) -> list[ChatCompletionOutput]:
+async def async_helper(client: CompletionEngine[Any], prompts: Prompts) -> list[ChatCompletionOutput]:
     return [result async for result in client.async_run(prompts)]
 
 
diff --git a/tests/test_text_to_speech_model.py b/tests/test_text_to_speech_model.py
index 89d8a88..dc44be4 100644
--- a/tests/test_text_to_speech_model.py
+++ b/tests/test_text_to_speech_model.py
@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 import asyncio
-from typing import Type
+from typing import Any, Type
 
 import pytest
 
@@ -19,7 +19,7 @@ def test_model_type_is_unique() -> None:
 
 
 @pytest.mark.parametrize('speech_model', get_pytest_params('test_text_to_speech', SpeechModelRegistry, types='model'))
-def test_speech_model(speech_model: TextToSpeechModel) -> None:
+def test_speech_model(speech_model: TextToSpeechModel[Any]) -> None:
     prompt = '你好，这是一个测试用例'
     sync_output = speech_model.generate(prompt)
     async_output = asyncio.run(speech_model.async_generate(prompt))
@@ -32,7 +32,7 @@ def test_speech_model(speech_model: TextToSpeechModel) -> None:
     ('model_cls', 'parameters'),
     get_pytest_params('test_speech_parameters', SpeechModelRegistry, types=('model_cls', 'parameter')),
 )
-def test_init_chat_parameters(model_cls: Type[TextToSpeechModel], parameters: ModelParameters) -> None:
-    parameters.voice_id = 'test'
+def test_init_chat_parameters(model_cls: Type[TextToSpeechModel[Any]], parameters: ModelParameters) -> None:
+    parameters.voice = 'test'
     model = model_cls(parameters=parameters)
-    assert model.parameters.voice_id == 'test'
+    assert model.parameters.voice == 'test'