Update app.py

app.py

@@ -1,9 +1,9 @@
-# 🚀 Main App - TalkingAIResearcher with Chat, Voice, Media, ArXiv, and More
 import streamlit as st
 import asyncio
 import websockets
 import uuid
 import argparse
+from datetime import datetime
 import os
 import random
 import time
@@ -17,6 +17,7 @@ import edge_tts
 from audio_recorder_streamlit import audio_recorder
 import nest_asyncio
 import re
+from streamlit_paste_button import paste_image_button
 import pytz
 import shutil
 import anthropic
@@ -28,11 +29,10 @@ import zipfile
 from gradio_client import Client
 from dotenv import load_dotenv
 from streamlit_marquee import streamlit_marquee
-from datetime import datetime
 from collections import defaultdict, Counter
 import pandas as pd
 
-# 🛠️ Patch asyncio for nesting glory
+# 🛠️ Patch asyncio for nesting
 nest_asyncio.apply()
 
 # 🎨 Page Config
@@ -57,8 +57,18 @@ FUN_USERNAMES = {
     "GalacticGopher 🌍": "en-AU-WilliamNeural",
     "RocketRaccoon 🚀": "en-CA-LiamNeural",
     "EchoElf 🧝": "en-US-AnaNeural",
+    "PhantomFox 🦊": "en-US-BrandonNeural",
+    "WittyWizard 🧙": "en-GB-ThomasNeural",
+    "LunarLlama 🌙": "en-AU-FreyaNeural",
+    "SolarSloth ☀️": "en-CA-LindaNeural",
+    "AstroAlpaca 🦙": "en-US-ChristopherNeural",
+    "CyberCoyote 🐺": "en-GB-ElliotNeural",
+    "MysticMoose 🦌": "en-AU-JamesNeural",
+    "GlitchGnome 🧚": "en-CA-EthanNeural",
+    "VortexViper 🐍": "en-US-AmberNeural",
+    "ChronoChimp 🐒": "en-GB-LibbyNeural"
 }
-EDGE_TTS_VOICES = list(set(FUN_USERNAMES.values()))  # 🎙️ Voice options
+EDGE_TTS_VOICES = list(set(FUN_USERNAMES.values()))
 FILE_EMOJIS = {"md": "📝", "mp3": "🎵", "wav": "🔊"}
 
 # 📁 Directories
@@ -69,7 +79,7 @@ CHAT_FILE = "chat_logs/global_chat.md"
 HISTORY_FILE = "history_logs/chat_history.md"
 MEDIA_DIR = "media_files"
 AUDIO_CACHE_DIR = "audio_cache"
-AUDIO_DIR = "audio_logs"  # New dir for MP3s
+AUDIO_DIR = "audio_logs"
 
 # 🔑 API Keys
 load_dotenv()
@@ -115,7 +125,6 @@ def init_session_state():
 
 # 🖌️ Marquee Helpers
 def update_marquee_settings_ui():
-    # 🎨 Sidebar marquee controls
     st.sidebar.markdown("### 🎯 Marquee Settings")
     cols = st.sidebar.columns(2)
     with cols[0]:
@@ -126,7 +135,6 @@ def update_marquee_settings_ui():
         st.session_state['marquee_settings']['animationDuration'] = f"{st.slider('⏱️ Speed', 1, 20, 20)}s"
 
 def display_marquee(text, settings, key_suffix=""):
-    # 🌈 Show marquee with truncation
     truncated = text[:280] + "..." if len(text) > 280 else text
     streamlit_marquee(content=truncated, **settings, key=f"marquee_{key_suffix}")
     st.write("")
@@ -142,7 +150,6 @@ def get_high_info_terms(text, top_n=10):
     return [t for t, _ in Counter(filtered).most_common(top_n)]
 
 def generate_filename(prompt, response, file_type="md"):
-    # 📁 Smart filename with info terms
     prefix = format_timestamp_prefix()
     terms = get_high_info_terms(prompt + " " + response, 5)
     snippet = clean_text_for_filename(prompt[:40] + " " + response[:40])
@@ -152,13 +159,11 @@ def generate_filename(prompt, response, file_type="md"):
     return f"{prefix}{base}_wct{wct}_sw{sw}_dur{dur}.{file_type}"
 
 def create_file(prompt, response, file_type="md"):
-    # 📝 Save file with Q&A
     filename = generate_filename(prompt, response, file_type)
     with open(filename, 'w', encoding='utf-8') as f: f.write(prompt + "\n\n" + response)
     return filename
 
 def get_download_link(file, file_type="mp3"):
-    # ⬇️ Cached download link
     cache_key = f"dl_{file}"
     if cache_key not in st.session_state['download_link_cache']:
         with open(file, "rb") as f:
@@ -168,7 +173,6 @@ def get_download_link(file, file_type="mp3"):
 
 # 🎶 Audio Processing
 async def async_edge_tts_generate(text, voice, username, rate=0, pitch=0, file_format="mp3"):
-    # 🎵 Async TTS with caching and .md generation
     cache_key = f"{text[:100]}_{voice}_{rate}_{pitch}_{file_format}"
     if cache_key in st.session_state['audio_cache']: return st.session_state['audio_cache'][cache_key], 0
     start_time = time.time()
@@ -179,7 +183,6 @@ async def async_edge_tts_generate(text, voice, username, rate=0, pitch=0, file_f
     await communicate.save(filename)
     st.session_state['audio_cache'][cache_key] = filename
     
-    # Generate .md file
     md_filename = filename.replace(".mp3", ".md")
     md_content = f"# Chat Audio Log\n\n**Player:** {username}\n**Voice:** {voice}\n**Text:**\n```markdown\n{text}\n```"
     with open(md_filename, 'w', encoding='utf-8') as f: f.write(md_content)
@@ -187,13 +190,11 @@ async def async_edge_tts_generate(text, voice, username, rate=0, pitch=0, file_f
     return filename, time.time() - start_time
 
 def play_and_download_audio(file_path):
-    # 🔊 Play + download
     if file_path and os.path.exists(file_path):
         st.audio(file_path)
         st.markdown(get_download_link(file_path), unsafe_allow_html=True)
 
 def load_mp3_viewer():
-    # 🎵 Load all MP3s at startup
     mp3_files = glob.glob(f"{AUDIO_DIR}/*.mp3")
     for mp3 in mp3_files:
         filename = os.path.basename(mp3)
@@ -201,7 +202,6 @@ def load_mp3_viewer():
             st.session_state['mp3_files'][filename] = mp3
 
 async def save_chat_entry(username, message, is_markdown=False):
-    # 💬 Save chat with multicast broadcast and audio
     central = pytz.timezone('US/Central')
     timestamp = datetime.now(central).strftime("%Y-%m-%d %H:%M:%S")
     entry = f"[{timestamp}] {username}: {message}" if not is_markdown else f"[{timestamp}] {username}:\n```markdown\n{message}\n```"
@@ -217,7 +217,6 @@ async def save_chat_entry(username, message, is_markdown=False):
     return audio_file
 
 async def load_chat():
-    # 📜 Load chat history - Numbered
     if not os.path.exists(CHAT_FILE):
         with open(CHAT_FILE, 'a') as f: f.write(f"# {START_ROOM} Chat\n\nWelcome to the cosmic hub! 🎤\n")
     with open(CHAT_FILE, 'r') as f: 
@@ -228,7 +227,6 @@ async def load_chat():
 
 # 🌐 WebSocket Handling
 async def websocket_handler(websocket, path):
-    # 🤝 Handle WebSocket clients
     client_id = str(uuid.uuid4())
     room_id = "chat"
     if room_id not in st.session_state.active_connections:
@@ -252,7 +250,6 @@ async def websocket_handler(websocket, path):
             del st.session_state.active_connections[room_id][client_id]
 
 async def broadcast_message(message, room_id):
-    # 📢 Broadcast to all clients
     if room_id in st.session_state.active_connections:
         disconnected = []
         for client_id, ws in st.session_state.active_connections[room_id].items():
@@ -265,7 +262,6 @@ async def broadcast_message(message, room_id):
                 del st.session_state.active_connections[room_id][client_id]
 
 async def run_websocket_server():
-    # 🖥️ Start WebSocket server
     if not st.session_state.server_running:
         server = await websockets.serve(websocket_handler, '0.0.0.0', 8765)
         st.session_state.server_running = True
@@ -282,7 +278,6 @@ class AudioProcessor:
         with open(f"{self.cache_dir}/metadata.json", 'w') as f: json.dump(self.metadata, f)
 
     async def create_audio(self, text, voice='en-US-AriaNeural'):
-        # 🎶 Generate cached audio
         cache_key = hashlib.md5(f"{text}:{voice}".encode()).hexdigest()
         cache_path = f"{self.cache_dir}/{cache_key}.mp3"
         if cache_key in self.metadata and os.path.exists(cache_path):
@@ -296,7 +291,6 @@ class AudioProcessor:
         return open(cache_path, 'rb').read()
 
 def process_pdf(pdf_file, max_pages, voice, audio_processor):
-    # 📄 Convert PDF to audio
     reader = PdfReader(pdf_file)
     total_pages = min(len(reader.pages), max_pages)
     texts, audios = [], {}
@@ -309,7 +303,6 @@ def process_pdf(pdf_file, max_pages, voice, audio_processor):
 
 # 🔍 ArXiv & AI Lookup
 def parse_arxiv_refs(ref_text):
-    # 📜 Parse ArXiv refs into dicts
     if not ref_text: return []
     papers = []
     current = {}
@@ -326,7 +319,6 @@ def parse_arxiv_refs(ref_text):
     return papers[:20]
 
 def generate_5min_feature_markdown(paper):
-    # ✨ 5-min research paper feature
     title, summary, authors, date, url = paper['title'], paper['summary'], paper['authors'], paper['date'], paper['url']
     pdf_url = url.replace("abs", "pdf") + (".pdf" if not url.endswith(".pdf") else "")
     wct, sw = len(title.split()), len(summary.split())
@@ -348,14 +340,12 @@ def generate_5min_feature_markdown(paper):
 def create_detailed_paper_md(papers): return "# Detailed Summary\n" + "\n".join(generate_5min_feature_markdown(p) for p in papers)
 
 async def create_paper_audio_files(papers, query):
-    # 🎧 Generate paper audio
     for p in papers:
         audio_text = clean_text_for_tts(f"{p['title']} by {p['authors']}. {p['summary']}")
         p['full_audio'], _ = await async_edge_tts_generate(audio_text, st.session_state['tts_voice'], p['authors'])
         if p['full_audio']: p['download_base64'] = get_download_link(p['full_audio'])
 
 async def perform_ai_lookup(q, useArxiv=True, useArxivAudio=False):
-    # 🔮 AI-powered research
     client = anthropic.Anthropic(api_key=anthropic_key)
     response = client.messages.create(model="claude-3-sonnet-20240229", max_tokens=1000, messages=[{"role": "user", "content": q}])
     result = response.content[0].text
@@ -378,7 +368,6 @@ async def perform_ai_lookup(q, useArxiv=True, useArxivAudio=False):
 
 # 📦 Zip Files
 def create_zip_of_files(md_files, mp3_files, query):
-    # 📦 Zip it up
     all_files = md_files + mp3_files
     if not all_files: return None
     terms = get_high_info_terms(" ".join([open(f, 'r', encoding='utf-8').read() if f.endswith('.md') else os.path.splitext(os.path.basename(f))[0].replace('_', ' ') for f in all_files] + [query]), 5)
@@ -389,7 +378,7 @@ def create_zip_of_files(md_files, mp3_files, query):
 # 🎮 Main Interface
 async def async_interface():
     init_session_state()
-    load_mp3_viewer()  # Load MP3s at startup
+    load_mp3_viewer()
     if not st.session_state.username:
         available = [n for n in FUN_USERNAMES if not any(f"{n} has joined" in l for l in (await load_chat()).split('\n'))]
         st.session_state.username = random.choice(available or list(FUN_USERNAMES.keys()))
@@ -418,7 +407,6 @@ async def async_interface():
             for i, line in enumerate(lines):
                 if line.strip():
                     st.markdown(line)
-                    # Match MP3 to line by timestamp and username
                     for mp3_name, mp3_path in st.session_state['mp3_files'].items():
                         if line.strip() in mp3_name and st.session_state.username in mp3_name:
                             st.audio(mp3_path, key=f"audio_{i}_{mp3_name}")
@@ -432,7 +420,7 @@ async def async_interface():
                 st.rerun()
 
         st.subheader("🎤 Speech-to-Chat")
-        speech_component = components.declare_component("speech_component", path="mycomponent")
+        from mycomponent import speech_component
         transcript_data = speech_component(default_value=st.session_state.get('last_transcript', ''))
         if transcript_data and 'value' in transcript_data:
             transcript = transcript_data['value'].strip()
@@ -498,7 +486,7 @@ async def async_interface():
                             st.audio(audios[i], format='audio/mp3')
                             st.markdown(get_download_link(io.BytesIO(audios[i]), "mp3"), unsafe_allow_html=True)
 
-    # 🗂️ Sidebar
+    # 🗂️ Sidebar with Dialog and Audio
     st.sidebar.subheader("Voice Settings")
     new_username = st.sidebar.selectbox("Change Name/Voice", list(FUN_USERNAMES.keys()), index=list(FUN_USERNAMES.keys()).index(st.session_state.username))
     if new_username != st.session_state.username:
@@ -506,6 +494,20 @@ async def async_interface():
         st.session_state.username, st.session_state.tts_voice = new_username, FUN_USERNAMES[new_username]
         st.rerun()
 
+    st.sidebar.markdown("### 💬 Chat Dialog & Audio")
+    chat_content = await load_chat()
+    lines = chat_content.split('\n')
+    audio_files = sorted(glob.glob(f"{AUDIO_DIR}/*.mp3"), key=os.path.getmtime, reverse=True)
+    for line in lines[-10:]:  # Show last 10 lines for brevity
+        if line.strip():
+            st.sidebar.markdown(f"**{line}**")
+            for mp3 in audio_files:
+                mp3_name = os.path.basename(mp3)
+                if st.session_state.username in mp3_name and any(word in mp3_name for word in line.split()):
+                    st.sidebar.audio(mp3, key=f"sidebar_audio_{mp3_name}")
+                    st.sidebar.markdown(get_download_link(mp3), unsafe_allow_html=True)
+                    break
+
     md_files, mp3_files = glob.glob("*.md"), glob.glob(f"{AUDIO_DIR}/*.mp3")
     st.sidebar.markdown("### 📂 File History")
     for f in sorted(md_files + mp3_files, key=os.path.getmtime, reverse=True)[:10]:
@@ -515,7 +517,6 @@ async def async_interface():
         if zip_name: st.sidebar.markdown(get_download_link(zip_name, "zip"), unsafe_allow_html=True)
 
 def main():
-    # 🎉 Kick it off
     asyncio.run(async_interface())
 
 if __name__ == "__main__":