diff --git a/rootfs/agents/studio_director/agent.toml b/rootfs/agents/studio_director/agent.toml
index 65a7c07..5a5d297 100644
--- a/rootfs/agents/studio_director/agent.toml
+++ b/rootfs/agents/studio_director/agent.toml
@@ -1,7 +1,8 @@
 instance_id = "studio_director"
 fullname = "tracy wang"
+llm_model_name = "gpt-3.5-turbo-16k-0613"
 enable_function = ["text_to_speech"]
 
 [[prompt]]
 role = "system"
-content = "你是一个故事播音员，请将下面故事改编成播音剧本，提取旁白和角色台词，以及每个角色需要有性别、年龄、以及每句台词的语气，最后生成音频文件。"
+content = "你是一个故事播音员，可以将故事演播成音频，演播前需要将故事改编成播音剧本，提取旁白和角色台词，以及每个角色需要有性别、年龄、以及每句台词的语气等。如果生成了音频文件则告知你的用户。"
diff --git a/rootfs/workflows/story_maker/workflow.toml b/rootfs/workflows/story_maker/workflow.toml
index 540665f..6ae5e55 100644
--- a/rootfs/workflows/story_maker/workflow.toml
+++ b/rootfs/workflows/story_maker/workflow.toml
@@ -16,7 +16,8 @@ content="""
 你是一个语音故事制作总导演,与客户对接并向团队下达指令。你的团队分为下面几个成员:writer,studio_director。一个故事制作分成两个阶段：让writer写出故事，再交由studio_director演播故事生成音频文件。你的基本工作模式是：
 1. 收到客户的明确的指令后,让writer写出故事
 2. 将writer写出的故事交给studio_director演播
-3. 当你决定要和成员通信时，请使用下面形式输出需要通信的消息
+3. 获得音频文件之后，将音频文件的存放路径交给客户
+4. 当你决定要和成员通信时，请使用下面形式输出需要通信的消息
 ```
 ##/send_msg 成员名称
 内容
diff --git a/src/aios_kernel/text_to_speech_function.py b/src/aios_kernel/text_to_speech_function.py
index 2f7bff0..90dfd49 100644
--- a/src/aios_kernel/text_to_speech_function.py
+++ b/src/aios_kernel/text_to_speech_function.py
@@ -14,7 +14,7 @@ logger = logging.getLogger(__name__)
 class TextToSpeechFunction(AIFunction):
     def __init__(self):
         self.func_id = "text_to_speech"
-        self.description = "根据输入的剧本生成音频数据"
+        self.description = "根据输入的剧本生成音频文件，成功时会返回音频文件路径"
 
     def get_name(self) -> str:
         return self.func_id
@@ -85,9 +85,9 @@ class TextToSpeechFunction(AIFunction):
         if audio is not None:
             path = os.path.join(os.path.realpath(os.curdir), "{}.mp3".format(''.join(random.sample('zyxwvutsrqponmlkjihgfedcba', 10))))
             audio.export(path, format="mp3")
-            return "已经生成音频文件, 文件路径为{}".format(path)
+            return "exec text_to_speech OK，speech file store at {}".format(path)
         else:
-            return "failed"
+            return "exec text_to_speech failed"
 
     def is_local(self) -> bool:
         return True