Update app.py

app.py

@@ -50,6 +50,44 @@ st.set_page_config(
     }
 )
 
+# 2. Load environment variables and initialize clients
+load_dotenv()
+
+# OpenAI setup
+openai.api_key = os.getenv('OPENAI_API_KEY')
+if openai.api_key == None:
+    openai.api_key = st.secrets['OPENAI_API_KEY']
+
+openai_client = OpenAI(
+    api_key=os.getenv('OPENAI_API_KEY'),
+    organization=os.getenv('OPENAI_ORG_ID')
+)
+
+# Claude setup
+anthropic_key = os.getenv("ANTHROPIC_API_KEY_3")
+if anthropic_key == None:
+    anthropic_key = st.secrets["ANTHROPIC_API_KEY"]
+claude_client = anthropic.Anthropic(api_key=anthropic_key)
+
+# HuggingFace setup
+API_URL = os.getenv('API_URL')
+HF_KEY = os.getenv('HF_KEY')
+MODEL1 = "meta-llama/Llama-2-7b-chat-hf"
+MODEL2 = "openai/whisper-small.en"
+
+headers = {
+    "Authorization": f"Bearer {HF_KEY}",
+    "Content-Type": "application/json"
+}
+
+# Initialize session states
+if "chat_history" not in st.session_state:
+    st.session_state.chat_history = []
+if "openai_model" not in st.session_state:
+    st.session_state["openai_model"] = "gpt-4o-2024-05-13"
+if "messages" not in st.session_state:
+    st.session_state.messages = []
+
 # Custom CSS
 st.markdown("""
 <style>
@@ -90,33 +128,6 @@ st.markdown("""
 </style>
 """, unsafe_allow_html=True)
 
-# 2. Load environment variables and initialize clients
-load_dotenv()
-
-# OpenAI setup
-openai.api_key = os.getenv('OPENAI_API_KEY')
-if openai.api_key == None:
-    openai.api_key = st.secrets['OPENAI_API_KEY']
-
-openai_client = OpenAI(
-    api_key=os.getenv('OPENAI_API_KEY'),
-    organization=os.getenv('OPENAI_ORG_ID')
-)
-
-# Claude setup
-anthropic_key = os.getenv("ANTHROPIC_API_KEY_3")
-if anthropic_key == None:
-    anthropic_key = st.secrets["ANTHROPIC_API_KEY"]
-claude_client = anthropic.Anthropic(api_key=anthropic_key)
-
-# Initialize session states
-if "chat_history" not in st.session_state:
-    st.session_state.chat_history = []
-if "openai_model" not in st.session_state:
-    st.session_state["openai_model"] = "gpt-4o-2024-05-13"
-if "messages" not in st.session_state:
-    st.session_state.messages = []
-
 # Bike Collections
 bike_collections = {
     "Celestial Collection 🌌": {
@@ -171,19 +182,27 @@ bike_collections = {
 
 # Helper Functions
 def generate_filename(prompt, file_type):
+    """Generate a safe filename using the prompt and file type."""
     central = pytz.timezone('US/Central')
     safe_date_time = datetime.now(central).strftime("%m%d_%H%M")
     replaced_prompt = re.sub(r'[<>:"/\\|?*\n]', ' ', prompt)
     safe_prompt = re.sub(r'\s+', ' ', replaced_prompt).strip()[:240]
     return f"{safe_date_time}_{safe_prompt}.{file_type}"
 
+
+
+
+# Function to create and save a file (and avoid the black hole of lost data 🕳)
 def create_file(filename, prompt, response, should_save=True):
     if not should_save:
         return
     with open(filename, 'w', encoding='utf-8') as file:
         file.write(prompt + "\n\n" + response)
 
+
+
 def create_and_save_file(content, file_type="md", prompt=None, is_image=False, should_save=True):
+    """Create and save file with proper handling of different types."""
     if not should_save:
         return None
     filename = generate_filename(prompt if prompt else content, file_type)
@@ -195,42 +214,41 @@ def create_and_save_file(content, file_type="md", prompt=None, is_image=False, s
     return filename
 
 def get_download_link(file_path):
+    """Create download link for file."""
     with open(file_path, "rb") as file:
         contents = file.read()
     b64 = base64.b64encode(contents).decode()
     return f'<a href="data:file/txt;base64,{b64}" download="{os.path.basename(file_path)}">Download {os.path.basename(file_path)}📂</a>'
 
-def load_file(file_name):
-    with open(file_name, "r", encoding='utf-8') as file:
-        content = file.read()
-    return content
-
-def create_zip_of_files(files):
-    zip_name = "all_files.zip"
-    with zipfile.ZipFile(zip_name, 'w') as zipf:
-        for file in files:
-            zipf.write(file)
-    return zip_name
-
-def get_media_html(media_path, media_type="video", width="100%"):
-    media_data = base64.b64encode(open(media_path, 'rb').read()).decode()
-    if media_type == "video":
-        return f'''
-        <video width="{width}" controls autoplay muted loop>
-            <source src="data:video/mp4;base64,{media_data}" type="video/mp4">
-            Your browser does not support the video tag.
-        </video>
-        '''
-    else:  # audio
-        return f'''
-        <audio controls style="width: {width};">
-            <source src="data:audio/mpeg;base64,{media_data}" type="audio/mpeg">
-            Your browser does not support the audio element.
-        </audio>
-        '''
+@st.cache_resource
+def SpeechSynthesis(result):
+    """HTML5 Speech Synthesis."""
+    documentHTML5 = f'''
+    <!DOCTYPE html>
+    <html>
+    <head>
+        <title>Read It Aloud</title>
+        <script type="text/javascript">
+            function readAloud() {{
+                const text = document.getElementById("textArea").value;
+                const speech = new SpeechSynthesisUtterance(text);
+                window.speechSynthesis.speak(speech);
+            }}
+        </script>
+    </head>
+    <body>
+        <h1>🔊 Read It Aloud</h1>
+        <textarea id="textArea" rows="10" cols="80">{result}</textarea>
+        <br>
+        <button onclick="readAloud()">🔊 Read Aloud</button>
+    </body>
+    </html>
+    '''
+    components.html(documentHTML5, width=1280, height=300)
 
 # Media Processing Functions
 def process_image(image_input, user_prompt):
+    """Process image with GPT-4o vision."""
     if isinstance(image_input, str):
         with open(image_input, "rb") as image_file:
             image_input = image_file.read()
@@ -254,6 +272,7 @@ def process_image(image_input, user_prompt):
     return response.choices[0].message.content
 
 def process_audio(audio_input, text_input=''):
+    """Process audio with Whisper and GPT."""
     if isinstance(audio_input, str):
         with open(audio_input, "rb") as file:
             audio_input = file.read()
@@ -267,11 +286,13 @@ def process_audio(audio_input, text_input=''):
     
     with st.chat_message("assistant"):
         st.markdown(transcription.text)
+        SpeechSynthesis(transcription.text)
         
         filename = generate_filename(transcription.text, "wav")
         create_and_save_file(audio_input, "wav", transcription.text, True)
 
 def process_video(video_path, seconds_per_frame=1):
+    """Process video files for frame extraction and audio."""
     base64Frames = []
     video = cv2.VideoCapture(video_path)
     total_frames = int(video.get(cv2.CAP_PROP_FRAME_COUNT))
@@ -302,6 +323,7 @@ def process_video(video_path, seconds_per_frame=1):
     return base64Frames, audio_path
 
 def process_video_with_gpt(video_input, user_prompt):
+    """Process video with GPT-4o vision."""
     base64Frames, audio_path = process_video(video_input)
     
     response = openai_client.chat.completions.create(
@@ -317,9 +339,104 @@ def process_video_with_gpt(video_input, user_prompt):
     )
     
     return response.choices[0].message.content
-    
-@st.cache_resource
+
+# ArXiv Search Functions
+def search_arxiv(query):
+    """Search ArXiv papers using Hugging Face client."""
+    client = Client("awacke1/Arxiv-Paper-Search-And-QA-RAG-Pattern")
+    response = client.predict(
+        query,
+        "mistralai/Mixtral-8x7B-Instruct-v0.1",
+        True,
+        api_name="/ask_llm"
+    )
+    return response
+
+# Chat Processing Functions
+def process_with_gpt(text_input):
+    """Process text with GPT-4o."""
+    if text_input:
+        st.session_state.messages.append({"role": "user", "content": text_input})
+        
+        with st.chat_message("user"):
+            st.markdown(text_input)
+        
+        with st.chat_message("assistant"):
+            completion = openai_client.chat.completions.create(
+                model=st.session_state["openai_model"],
+                messages=[
+                    {"role": m["role"], "content": m["content"]}
+                    for m in st.session_state.messages
+                ],
+                stream=False
+            )
+            return_text = completion.choices[0].message.content
+            st.write("GPT-4o: " + return_text)
+            
+            filename = generate_filename(text_input, "md")
+            create_file(filename, text_input, return_text)
+            st.session_state.messages.append({"role": "assistant", "content": return_text})
+            return return_text
+
+def process_with_claude(text_input):
+    """Process text with Claude."""
+    if text_input:
+        response = claude_client.messages.create(
+            model="claude-3-sonnet-20240229",
+            max_tokens=1000,
+            messages=[
+                {"role": "user", "content": text_input}
+            ]
+        )
+        response_text = response.content[0].text
+        st.write("Claude: " + response_text)
+        
+        filename = generate_filename(text_input, "md")
+        create_file(filename, text_input, response_text)
+        
+        st.session_state.chat_history.append({
+            "user": text_input,
+            "claude": response_text
+        })
+        return response_text
+
+# File Management Functions
+def load_file(file_name):
+    """Load file content."""
+    with open(file_name, "r", encoding='utf-8') as file:
+        content = file.read()
+    return content
+
+def create_zip_of_files(files):
+    """Create zip archive of files."""
+    zip_name = "all_files.zip"
+    with zipfile.ZipFile(zip_name, 'w') as zipf:
+        for file in files:
+            zipf.write(file)
+    return zip_name
+
+
+
+def get_media_html(media_path, media_type="video", width="100%"):
+    """Generate HTML for media player."""
+    media_data = base64.b64encode(open(media_path, 'rb').read()).decode()
+    if media_type == "video":
+        return f'''
+        <video width="{width}" controls autoplay muted loop>
+            <source src="data:video/mp4;base64,{media_data}" type="video/mp4">
+            Your browser does not support the video tag.
+        </video>
+        '''
+    else:  # audio
+        return f'''
+        <audio controls style="width: {width};">
+            <source src="data:audio/mpeg;base64,{media_data}" type="audio/mpeg">
+            Your browser does not support the audio element.
+        </audio>
+        '''
+
 def create_media_gallery():
+    """Create the media gallery interface."""
     st.header("🎬 Media Gallery")
     
     tabs = st.tabs(["🖼️ Images", "🎵 Audio", "🎥 Video", "🎨 Scene Generator"])