diff --git a/tests/integration_tests/test_audio.py b/tests/integration_tests/test_audio.py
index 2ef05d6..e4891b5 100644
--- a/tests/integration_tests/test_audio.py
+++ b/tests/integration_tests/test_audio.py
@@ -1,5 +1,3 @@
-import base64
-import json
 import logging
 import logging.config
 from pathlib import Path
@@ -16,8 +14,7 @@ def test_audio_speech(logging_conf):
 		response = client.audio.speech(
 			model='cogtts',
 			input='你好呀,欢迎来到智谱开放平台',
-			voice='tongtong',
-			stream=False,
+			voice='female',
 			response_format='wav',
 		)
 		response.stream_to_file(speech_file_path)
@@ -29,37 +26,6 @@ def test_audio_speech(logging_conf):
 	except zhipuai.core._errors.APIStatusError as err:
 		print(err)
 
-def test_audio_speech_streaming(logging_conf):
-	logging.config.dictConfig(logging_conf)  # type: ignore
-	client = ZhipuAI()  # 填写您自己的APIKey
-	try:
-		response = client.audio.speech(
-			model='cogtts',
-			input='你好呀,欢迎来到智谱开放平台',
-			voice='tongtong',
-			stream=True,
-			response_format='wav',
-		)
-		with open("output.pcm", "wb") as f:
-			for item in response:
-				choice = item.choices[0]
-				index = choice.index
-				finish_reason = choice.finish_reason
-				audio_delta = choice.delta.content
-				if finish_reason is not None:
-					break
-				f.write(base64.b64decode(audio_delta))
-				print(f"{index}.finish_reason = {finish_reason}, audio_delta = {len(audio_delta)}")
-
-	except zhipuai.core._errors.APIRequestFailedError as err:
-		print(err)
-	except zhipuai.core._errors.APIInternalError as err:
-		print(err)
-	except zhipuai.core._errors.APIStatusError as err:
-		print(err)
-	except Exception as e:
-		print(e)
-
 
 def test_audio_customization(logging_conf):
 	logging.config.dictConfig(logging_conf)
diff --git a/zhipuai/api_resource/audio/audio.py b/zhipuai/api_resource/audio/audio.py
index d916048..aa987cb 100644
--- a/zhipuai/api_resource/audio/audio.py
+++ b/zhipuai/api_resource/audio/audio.py
@@ -9,7 +9,7 @@
 from zhipuai.types.audio import AudioSpeechParams
 from ...types.audio import audio_customization_param
 
-from zhipuai.core import BaseAPI, maybe_transform, StreamResponse
+from zhipuai.core import BaseAPI, maybe_transform
 from zhipuai.core import NOT_GIVEN, Body, Headers, NotGiven, FileTypes
 from zhipuai.core import _legacy_response
 
@@ -20,7 +20,6 @@
     make_request_options,
 )
 from zhipuai.core import deepcopy_minimal
-from ...types.audio.audio_speech_chunk import AudioSpeechChunk
 
 if TYPE_CHECKING:
     from zhipuai._client import ZhipuAI
@@ -47,17 +46,15 @@ def speech(
             sensitive_word_check: Optional[SensitiveWordCheckRequest] | NotGiven = NOT_GIVEN,
             request_id: str = None,
             user_id: str = None,
-            stream: bool = False,
             extra_headers: Headers | None = None,
             extra_body: Body | None = None,
             timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
-    ) -> _legacy_response.HttpxBinaryResponseContent | StreamResponse[AudioSpeechChunk]:
+    ) -> _legacy_response.HttpxBinaryResponseContent:
         body = deepcopy_minimal(
             {
                 "model": model,
                 "input": input,
                 "voice": voice,
-                "stream": stream,
                 "response_format": response_format,
                 "sensitive_word_check": sensitive_word_check,
                 "request_id": request_id,
@@ -66,13 +63,11 @@ def speech(
         )
         return self._post(
             "/audio/speech",
-            body=body,
+            body=maybe_transform(body, AudioSpeechParams),
             options=make_request_options(
                 extra_headers=extra_headers, extra_body=extra_body, timeout=timeout
             ),
-            cast_type=_legacy_response.HttpxBinaryResponseContent,
-            stream= stream or False,
-            stream_cls=StreamResponse[AudioSpeechChunk]
+            cast_type=_legacy_response.HttpxBinaryResponseContent
         )
 
     def customization(
diff --git a/zhipuai/types/audio/audio_speech_chunk.py b/zhipuai/types/audio/audio_speech_chunk.py
deleted file mode 100644
index 7788d9d..0000000
--- a/zhipuai/types/audio/audio_speech_chunk.py
+++ /dev/null
@@ -1,32 +0,0 @@
-from typing import List, Optional, Dict, Any
-
-from ...core import BaseModel
-
-__all__ = [
-    "AudioSpeechChunk",
-    "AudioError",
-    "AudioSpeechChoice",
-    "AudioSpeechDelta"
-]
-
-
-class AudioSpeechDelta(BaseModel):
-    content: Optional[str] = None
-    role: Optional[str] = None
-
-
-class AudioSpeechChoice(BaseModel):
-    delta: AudioSpeechDelta
-    finish_reason: Optional[str] = None
-    index: int
-
-class AudioError:
-    code: Optional[str] = None
-    message: Optional[str] = None
-
-
-class AudioSpeechChunk(BaseModel):
-    choices: List[AudioSpeechChoice]
-    request_id: Optional[str] = None
-    created: Optional[int] = None
-    error: Optional[AudioError] = None
\ No newline at end of file