chore: update version to 2.1.5.a1 and refactor AudioError class

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "zhipuai"
-version = "2.1.5.20250724"
+version = "2.1.5.a1"
 description = "A SDK library for accessing big model apis from ZhipuAI"
 authors = ["Zhipu AI"]
 readme = "README.md"

tests/unit_tests/test_audio.py

@@ -0,0 +1,30 @@
+def test_audio_error_field():
+    from zhipuai.types.audio.audio_speech_chunk import AudioSpeechChunk, AudioError, AudioSpeechChoice, AudioSpeechDelta
+
+    # 构造一个 AudioError
+    error = AudioError(code="500", message="Internal Error")
+
+    # 构造一个完整的 AudioSpeechChunk
+    chunk = AudioSpeechChunk(
+        choices=[
+            AudioSpeechChoice(
+                delta=AudioSpeechDelta(content="audio", role="system"),
+                finish_reason="error",
+                index=0
+            )
+        ],
+        request_id="req_2",
+        created=123456,
+        error=error
+    )
+
+    # 检查 error 字段是否为 AudioError 实例
+    assert isinstance(chunk.error, AudioError)
+    assert chunk.error.code == "500"
+    assert chunk.error.message == "Internal Error"
+
+    # 检查序列化
+    as_dict = chunk.model_dump()
+    assert as_dict["error"]["code"] == "500"
+    assert as_dict["error"]["message"] == "Internal Error"
+    print("test_audio_error_field passed.")

zhipuai/__version__.py

@@ -1 +1 @@
-__version__ = 'v2.1.5.20250724'
+__version__ = 'v2.1.5.a1'

zhipuai/types/audio/audio_speech_chunk.py

@@ -20,7 +20,7 @@ class AudioSpeechChoice(BaseModel):
     finish_reason: Optional[str] = None
     index: int
 
-class AudioError:
+class AudioError(BaseModel):
     code: Optional[str] = None
     message: Optional[str] = None