Add application file

.env

@@ -0,0 +1 @@
+OPENAI_API_KEY="sk-ut4AJ8UzqsoTVaeIATrXBGfLRcYhZbba04jnSZyOwlT3BlbkFJsEa4UxwYuflmBYBi2oLQnZJIB8-zCzTWJ9QQLMzt8A"

app.py

@@ -3,13 +3,22 @@ from PIL import Image
 from io import BytesIO
 import requests
 import time
-from openai import OpenAI, AzureOpenAI
+import openai
+import os
+from openai import OpenAI
+from dotenv import load_dotenv
 
-engine = "gpt-4o"
-api_key = "9e95c7efd64648bd86cd2fcfbe17905b"
-api_base = "https://chatpartner-miniprogram.openai.azure.com/"
-api_version = "2024-02-15-preview"
-client = AzureOpenAI(api_key=api_key, azure_endpoint=api_base, api_version=api_version)
+load_dotenv()
+
+client = OpenAI(
+    # This is the default and can be omitted
+    api_key = os.getenv("OPENAI_API_KEY")
+    #api_key=os.environ.get("OPENAI_API_KEY"),
+)
+
+# 设置OpenAI API密钥
+openai.api_key = "sk-ut4AJ8UzqsoTVaeIATrXBGfLRcYhZbba04jnSZyOwlT3BlbkFJsEa4UxwYuflmBYBi2oLQnZJIB8-zCzTWJ9QQLMzt8A"
+engine = "gpt-4o-mini"
 
 class ImageProcessor:
     def __init__(self, api_key):
@@ -60,7 +69,7 @@ class ImageProcessor:
         task_id = self.submit_images(image_bytes_list)
         return self.get_result(task_id)
 
-# 设置API密钥
+# 设置Music API密钥
 api_key = 'ddc85b14-bd83-4757-9bc4-8a11194da536'
 image_processor = ImageProcessor(api_key)
 
@@ -78,7 +87,7 @@ def process_input(text=None, images=None, audio=None):
         # 使用ImageProcessor处理图像
         image_bytes_list = []
         for image in images:
-            img = Image.open(image)
+            img = Image.open(image.name)
             image_bytes = BytesIO()
             img.save(image_bytes, format="PNG")
             image_bytes.seek(0)
@@ -86,7 +95,9 @@ def process_input(text=None, images=None, audio=None):
         
         try:
             processed_image_result = image_processor.process_images(image_bytes_list)
+            #prompt += f"\n乐谱的内容如下,这是一首杜维诺伊的曲子，请你根据他的曲风回答问题: {processed_image_result}"
             prompt += f"\n乐谱的内容如下,这是一首杜维诺伊的曲子，请你根据他的曲风回答问题: {processed_image_result}"
+        
         except Exception as e:
             return f"Error processing image: {e}", None
     
@@ -96,21 +107,11 @@ def process_input(text=None, images=None, audio=None):
     
     # 使用GPT API进行处理
     try:
-        '''
-        from zhipuai import ZhipuAI
-        client = ZhipuAI(api_key="7cfa094b2c6867229659831340af9f2c.BIl47duC5fufSY8V") # 填写您自己的APIKey
-        response = client.chat.completions.create(
-            model="glm-4",  # 填写需要调用的模型名称
-            messages=[
-            {"role": "user", "content": prompt},        
-            ],
-        )
-        '''
         messages.append({"role": "user", "content": prompt})
         response = client.chat.completions.create(
             model=engine,
             messages=messages,
-            temperature=0.1,
+            temperature=0.2,
             max_tokens=4096,
             top_p=0.95,
             frequency_penalty=0,
@@ -118,7 +119,7 @@ def process_input(text=None, images=None, audio=None):
             stop=None
         )
         print(response.choices[0].message.content)
-        return response.choices[0].message.content.strip()
+        return response.choices[0].message.content
     except Exception as e:
         return f"Error: {e}", None
 
@@ -126,17 +127,17 @@ def process_input(text=None, images=None, audio=None):
 iface = gr.Interface(
     fn=process_input,
     inputs=[
-        gr.Textbox(label="Input Text", placeholder="Enter text here"),
-        gr.File(label="Input Images", file_count="multiple", type="filepath"),  # 支持多文件上传
+        gr.Textbox(label="Input Text", placeholder="Enter text here", lines=2),  # Text input with submit button
+        gr.File(label="Input Images", file_count="multiple", type="file"),  # 支持多文件上传
         # gr.Audio(label="Input Audio", type="filepath"),
+        
     ],
     outputs=[
         gr.Textbox(label="Output Text"),
         # gr.Audio(label="Output Audio")  # 目前示例中未处理音频输出
     ],
-    live=True
+    live=False,
 )
 
 # 启动Gradio应用
 iface.launch()
-

requirements.txt

@@ -1,3 +1,4 @@
 openai
 gradio
 zhipuai
+python-dotenv