feat: 20250801 tts mp3 (#101)

Co-authored-by: yuhongxiao <[email protected]>

Author: iceAndFireisFailed

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "zhipuai"
-version = "2.1.5.20250726"
+version = "2.1.5.20250801"
 description = "A SDK library for accessing big model apis from ZhipuAI"
 authors = ["Zhipu AI"]
 readme = "README.md"

tests/integration_tests/test_audio.py

@@ -30,35 +30,35 @@ def test_audio_speech(logging_conf):
 		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)
+    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='mp3',
+          encode_format='hex'
+       )
+       with open("output.mp3", "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(bytes.fromhex(audio_delta))
+             print(f"audio delta: {audio_delta[:64]}..., 长度:{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):

zhipuai/api_resource/audio/audio.py

@@ -51,6 +51,7 @@ def speech(
             extra_headers: Headers | None = None,
             extra_body: Body | None = None,
             timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+            encode_format: str,
     ) -> _legacy_response.HttpxBinaryResponseContent | StreamResponse[AudioSpeechChunk]:
         body = deepcopy_minimal(
             {
@@ -61,7 +62,8 @@ def speech(
                 "response_format": response_format,
                 "sensitive_word_check": sensitive_word_check,
                 "request_id": request_id,
-                "user_id": user_id
+                "user_id": user_id,
+                "encode_format": encode_format
             }
         )
         return self._post(