Update app.py

app.py

@@ -34,8 +34,8 @@ from xml.etree import ElementTree as ET
 import streamlit.components.v1 as components  # Import Streamlit Components for HTML5
 
 
-st.set_page_config(page_title="🐪Llama🦙Whisperer", layout="wide")
-st.markdown('(Inference Endpoints)[https://ui.endpoints.huggingface.co/awacke1/endpoints]')
+st.set_page_config(page_title="🐪Llama Whisperer🦙 Voice Chat🌟", layout="wide")
+
 
 def add_Med_Licensing_Exam_Dataset():
     import streamlit as st
@@ -92,9 +92,9 @@ def add_Med_Licensing_Exam_Dataset():
 # 1. Constants and Top Level UI Variables
 
 # My Inference API Copy
-API_URL = 'https://qe55p8afio98s0u3.us-east-1.aws.endpoints.huggingface.cloud'  # Dr Llama
+# API_URL = 'https://qe55p8afio98s0u3.us-east-1.aws.endpoints.huggingface.cloud'  # Dr Llama
 # Original:
-#API_URL = "https://api-inference.huggingface.co/models/meta-llama/Llama-2-7b-chat-hf"
+API_URL = "https://api-inference.huggingface.co/models/meta-llama/Llama-2-7b-chat-hf"
 API_KEY = os.getenv('API_KEY')
 MODEL1="meta-llama/Llama-2-7b-chat-hf"
 MODEL1URL="https://huggingface.co/meta-llama/Llama-2-7b-chat-hf"
@@ -239,7 +239,8 @@ def generate_filename(prompt, file_type):
     central = pytz.timezone('US/Central')
     safe_date_time = datetime.now(central).strftime("%m%d_%H%M")
     replaced_prompt = prompt.replace(" ", "_").replace("\n", "_")
-    safe_prompt = "".join(x for x in replaced_prompt if x.isalnum() or x == "_")[:45]
+    safe_prompt = "".join(x for x in replaced_prompt if x.isalnum() or x == "_")[:255]  # 255 is linux max, 260 is windows max
+    #safe_prompt = "".join(x for x in replaced_prompt if x.isalnum() or x == "_")[:45]
     return f"{safe_date_time}_{safe_prompt}.{file_type}"
 
 # 6. Speech transcription via OpenAI service
@@ -326,6 +327,8 @@ def get_table_download_link(file_path):
         mime_type = 'text/html'
     elif ext == '.md':
         mime_type = 'text/markdown'
+    elif ext == '.wav':
+        mime_type = 'audio/wav'
     else:
         mime_type = 'application/octet-stream'  # general binary data type
     href = f'<a href="data:{mime_type};base64,{b64}" target="_blank" download="{file_name}">{file_name}</a>'
@@ -506,19 +509,17 @@ def get_zip_download_link(zip_file):
 
 # 14. Inference Endpoints for Whisper (best fastest STT) on NVIDIA T4 and Llama (best fastest AGI LLM) on NVIDIA A10
 # My Inference Endpoint
-#API_URL_IE = f'https://tonpixzfvq3791u9.us-east-1.aws.endpoints.huggingface.cloud'
+API_URL_IE = f'https://tonpixzfvq3791u9.us-east-1.aws.endpoints.huggingface.cloud'
 # Original
-#API_URL_IE = "https://api-inference.huggingface.co/models/openai/whisper-small.en"
-# A10 Inference Endpoint for whisper large tests
-API_URL_IE = "https://hifdvffh2em0wn50.us-east-1.aws.endpoints.huggingface.cloud"
-
+API_URL_IE = "https://api-inference.huggingface.co/models/openai/whisper-small.en"
 MODEL2 = "openai/whisper-small.en"
 MODEL2_URL = "https://huggingface.co/openai/whisper-small.en"
 #headers = {
 #	"Authorization": "Bearer XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX",
 #	"Content-Type": "audio/wav"
 #}
-HF_KEY = os.getenv('HF_KEY')
+# HF_KEY = os.getenv('HF_KEY')
+HF_KEY = st.secrets['HF_KEY']
 headers = {
     "Authorization": f"Bearer {HF_KEY}",
     "Content-Type": "audio/wav"
@@ -553,27 +554,28 @@ def transcribe_audio(filename):
     output = query(filename)
     return output
 
-
 def whisper_main():
-    st.title("1🐪Llama🦙Whisperer")
-    st.write("Record your speech and get the text.")
+    #st.title("Speech to Text")
+    #st.write("Record your speech and get the text.")
 
     # Audio, transcribe, GPT:
     filename = save_and_play_audio(audio_recorder)
     if filename is not None:
         transcription = transcribe_audio(filename)
-        #try:
-        
-        transcript = transcription['text']
-        #except:
-        #st.write('Whisper model is asleep. Starting up now on T4 GPU - please give 5 minutes then retry as it scales up from zero to activate running container(s).')
+        try:
+            transcript = transcription['text']
+            st.write(transcript)
+            response = StreamLLMChatResponse(transcript)
+            filename_txt = generate_filename(transcript, ".txt")
+            create_file(filename_txt, transcript, response, should_save)
+            filename_wav = filename_txt.replace('.txt', '.wav')
+            import shutil
+            shutil.copyfile(filename, filename_wav)
+            if os.path.exists(filename):
+                os.remove(filename)
+        except:
+            st.write('Starting Whisper Model on GPU.  Please retry in 30 seconds.')
 
-        st.write(transcript)
-        response = StreamLLMChatResponse(transcript)
-        # st.write(response) - redundant with streaming result?
-        filename = generate_filename(transcript, ".txt")
-        create_file(filename, transcript, response, should_save)
-        #st.sidebar.markdown(get_table_download_link(filename), unsafe_allow_html=True)
 
 import streamlit as st
 
@@ -581,130 +583,121 @@ import streamlit as st
 def StreamMedChatResponse(topic):
     st.write(f"Showing resources or questions related to: {topic}")
 
-def add_multi_system_agent_topics():
-    with st.expander("Multi-System Agent AI Topics 🤖", expanded=True):
-        st.markdown("🤖 **Explore Multi-System Agent AI Topics**: This section provides a variety of topics related to multi-system agent AI systems.")
+def add_medical_exam_buttons():
+    with st.expander("Medical Licensing Exam Topics 📚", expanded=False):
+        st.markdown("🩺 **Important**: This section provides a variety of medical topics that are often encountered in medical licensing exams.")
 
-        # Define multi-system agent AI topics and descriptions
+        # Define medical exam terminology descriptions
         descriptions = {
-            "Reinforcement Learning 🎮": "Questions related to reinforcement learning algorithms and applications 🕹️",
-            "Natural Language Processing 🗣️": "Questions about natural language processing techniques and chatbot development 🗨️",
-            "Multi-Agent Systems 🤝": "Questions pertaining to multi-agent systems and cooperative AI interactions 🤖",
-            "Conversational AI 🗨️": "Questions on building conversational AI agents and chatbots for various platforms 💬",
-            "Distributed AI Systems 🌐": "Questions about distributed AI systems and their implementation in networked environments 🌐",
-            "AI Ethics and Bias 🤔": "Questions related to ethics and bias considerations in AI systems and decision-making 🧠",
-            "AI in Healthcare 🏥": "Questions about the application of AI in healthcare and medical diagnosis 🩺",
-            "AI in Autonomous Vehicles 🚗": "Questions on the use of AI in autonomous vehicles and self-driving technology 🚗"
+            "Ultrasound with Doppler 🌊": "3 Questions and Answers with emojis about Doppler Ultrasound imaging techniques 🎥",
+            "Oseltamivir 🦠": "3 Questions and Answers with emojis about the antiviral medication Oseltamivir 💊",
+            "IM Epinephrine 💉": "3 Questions and Answers with emojis about intramuscular administration of epinephrine 💪",
+            "Hypokalemia 🍌": "3 Questions and Answers with emojis about low potassium levels in blood 🩸",
+            "Succinylcholine 💊": "3 Questions and Answers with emojis on the use and side-effects of Succinylcholine 🚑",
+            "Phosphoinositol System 🧬": "3 Questions and Answers with emojis  about the Phosphoinositol signalling system 🛠",
+            "Ramipril 💊": "3 Questions and Answers with emojis related to the ACE inhibitor Ramipril 🩺"
         }
 
         # Create columns
         col1, col2, col3, col4 = st.columns([1, 1, 1, 1], gap="small")
         
         # Add buttons to columns
-        if col1.button("Reinforcement Learning 🎮"):
-            st.write(descriptions["Reinforcement Learning 🎮"])
-            StreamLLMChatResponse(descriptions["Reinforcement Learning 🎮"])
+        if col1.button("Ultrasound with Doppler 🌊"):
+            StreamLLMChatResponse(descriptions["Ultrasound with Doppler 🌊"])
 
-        if col2.button("Natural Language Processing 🗣️"):
-            st.write(descriptions["Natural Language Processing 🗣️"])
-            StreamLLMChatResponse(descriptions["Natural Language Processing 🗣️"])
+        if col2.button("Oseltamivir 🦠"):
+            StreamLLMChatResponse(descriptions["Oseltamivir 🦠"])
 
-        if col3.button("Multi-Agent Systems 🤝"):
-            st.write(descriptions["Multi-Agent Systems 🤝"])
-            StreamLLMChatResponse(descriptions["Multi-Agent Systems 🤝"])
+        if col3.button("IM Epinephrine 💉"):
+            StreamLLMChatResponse(descriptions["IM Epinephrine 💉"])
 
-        if col4.button("Conversational AI 🗨️"):
-            st.write(descriptions["Conversational AI 🗨️"])
-            StreamLLMChatResponse(descriptions["Conversational AI 🗨️"])
+        if col4.button("Hypokalemia 🍌"):
+            StreamLLMChatResponse(descriptions["Hypokalemia 🍌"])
 
         col5, col6, col7, col8 = st.columns([1, 1, 1, 1], gap="small")
 
-        if col5.button("Distributed AI Systems 🌐"):
-            st.write(descriptions["Distributed AI Systems 🌐"])
-            StreamLLMChatResponse(descriptions["Distributed AI Systems 🌐"])
-
-        if col6.button("AI Ethics and Bias 🤔"):
-            st.write(descriptions["AI Ethics and Bias 🤔"])
-            StreamLLMChatResponse(descriptions["AI Ethics and Bias 🤔"])
-
-        if col7.button("AI in Healthcare 🏥"):
-            st.write(descriptions["AI in Healthcare 🏥"])
-            StreamLLMChatResponse(descriptions["AI in Healthcare 🏥"])
+        if col5.button("Succinylcholine 💊"):
+            StreamLLMChatResponse(descriptions["Succinylcholine 💊"])
 
-        if col8.button("AI in Autonomous Vehicles 🚗"):
-            st.write(descriptions["AI in Autonomous Vehicles 🚗"])
-            StreamLLMChatResponse(descriptions["AI in Autonomous Vehicles 🚗"])
+        if col6.button("Phosphoinositol System 🧬"):
+            StreamLLMChatResponse(descriptions["Phosphoinositol System 🧬"])
 
+        if col7.button("Ramipril 💊"):
+            StreamLLMChatResponse(descriptions["Ramipril 💊"])
+        
+        
 
 # 17. Main
 def main():
 
-    st.title("Try Some Topics:")
+    #st.title("GAIA - Medical License Exam Testing")
     prompt = f"Write ten funny jokes that are tweet length stories that make you laugh.  Show as markdown outline with emojis for each."
 
     # Add Wit and Humor buttons
     # add_witty_humor_buttons()
-    # Calling the function to add the multi-system agent AI topics buttons
-    add_multi_system_agent_topics()
+    add_medical_exam_buttons()
 
-    example_input = st.text_input("Enter your example text:", value=prompt, help="Enter text to get a response from DromeLlama.")
-    if st.button("Run Prompt With DromeLlama", help="Click to run the prompt."):
-        try:
-            StreamLLMChatResponse(example_input)
-        except:
-            st.write('DromeLlama is asleep. Starting up now on A10 - please give 5 minutes then retry as KEDA scales up from zero to activate running container(s).')
-
-    openai.api_key = os.getenv('OPENAI_KEY')
-    menu = ["txt", "htm", "xlsx", "csv", "md", "py"]
-    choice = st.sidebar.selectbox("Output File Type:", menu)
-    model_choice = st.sidebar.radio("Select Model:", ('gpt-3.5-turbo', 'gpt-3.5-turbo-0301'))        
-    user_prompt = st.text_area("Enter prompts, instructions & questions:", '', height=100)
-    collength, colupload = st.columns([2,3])  # adjust the ratio as needed
-    with collength:
-        max_length = st.slider("File section length for large files", min_value=1000, max_value=128000, value=12000, step=1000)
-    with colupload:
-        uploaded_file = st.file_uploader("Add a file for context:", type=["pdf", "xml", "json", "xlsx", "csv", "html", "htm", "md", "txt"])
-    document_sections = deque()
-    document_responses = {}
-    if uploaded_file is not None:
-        file_content = read_file_content(uploaded_file, max_length)
-        document_sections.extend(divide_document(file_content, max_length))
-    if len(document_sections) > 0:
-        if st.button("👁️ View Upload"):
-            st.markdown("**Sections of the uploaded file:**")
-            for i, section in enumerate(list(document_sections)):
-                st.markdown(f"**Section {i+1}**\n{section}")
-        st.markdown("**Chat with the model:**")
-        for i, section in enumerate(list(document_sections)):
-            if i in document_responses:
-                st.markdown(f"**Section {i+1}**\n{document_responses[i]}")
-            else:
-                if st.button(f"Chat about Section {i+1}"):
-                    st.write('Reasoning with your inputs...')
-                    response = chat_with_model(user_prompt, section, model_choice)
-                    st.write('Response:')
-                    st.write(response)
-                    document_responses[i] = response
-                    filename = generate_filename(f"{user_prompt}_section_{i+1}", choice)
-                    create_file(filename, user_prompt, response, should_save)
-                    st.sidebar.markdown(get_table_download_link(filename), unsafe_allow_html=True)
-    if st.button('💬 Chat'):
-        st.write('Reasoning with your inputs...')
-        user_prompt_sections = divide_prompt(user_prompt, max_length)
-        full_response = ''
-        for prompt_section in user_prompt_sections:
-            response = chat_with_model(prompt_section, ''.join(list(document_sections)), model_choice)
-            full_response += response + '\n'  # Combine the responses
-        response = full_response
-        st.write('Response:')
-        st.write(response)
-        filename = generate_filename(user_prompt, choice)
-        create_file(filename, user_prompt, response, should_save)
-        st.sidebar.markdown(get_table_download_link(filename), unsafe_allow_html=True)
 
-    # Compose a file sidebar of past encounters
-    all_files = glob.glob("*.*")
-    all_files = [file for file in all_files if len(os.path.splitext(file)[0]) >= 20]  # exclude files with short names
+    with st.expander("Prompts 📚", expanded=False):
+
+        example_input = st.text_input("Enter your example text:", value=prompt, help="Enter text to get a response from DromeLlama.")
+        if st.button("Run Prompt With DromeLlama", help="Click to run the prompt."):
+            try:
+                StreamLLMChatResponse(example_input)
+            except:
+                st.write('DromeLlama is asleep. Starting up now on A10 - please give 5 minutes then retry as KEDA scales up from zero to activate running container(s).')
+
+        openai.api_key = os.getenv('OPENAI_KEY')
+        menu = ["txt", "htm", "xlsx", "csv", "md", "py"]
+        choice = st.sidebar.selectbox("Output File Type:", menu)
+        model_choice = st.sidebar.radio("Select Model:", ('gpt-3.5-turbo', 'gpt-3.5-turbo-0301'))        
+        user_prompt = st.text_area("Enter prompts, instructions & questions:", '', height=100)
+        collength, colupload = st.columns([2,3])  # adjust the ratio as needed
+        with collength:
+            max_length = st.slider("File section length for large files", min_value=1000, max_value=128000, value=12000, step=1000)
+        with colupload:
+            uploaded_file = st.file_uploader("Add a file for context:", type=["pdf", "xml", "json", "xlsx", "csv", "html", "htm", "md", "txt"])
+        document_sections = deque()
+        document_responses = {}
+        if uploaded_file is not None:
+            file_content = read_file_content(uploaded_file, max_length)
+            document_sections.extend(divide_document(file_content, max_length))
+        if len(document_sections) > 0:
+            if st.button("👁️ View Upload"):
+                st.markdown("**Sections of the uploaded file:**")
+                for i, section in enumerate(list(document_sections)):
+                    st.markdown(f"**Section {i+1}**\n{section}")
+            st.markdown("**Chat with the model:**")
+            for i, section in enumerate(list(document_sections)):
+                if i in document_responses:
+                    st.markdown(f"**Section {i+1}**\n{document_responses[i]}")
+                else:
+                    if st.button(f"Chat about Section {i+1}"):
+                        st.write('Reasoning with your inputs...')
+                        response = chat_with_model(user_prompt, section, model_choice)
+                        st.write('Response:')
+                        st.write(response)
+                        document_responses[i] = response
+                        filename = generate_filename(f"{user_prompt}_section_{i+1}", choice)
+                        create_file(filename, user_prompt, response, should_save)
+                        st.sidebar.markdown(get_table_download_link(filename), unsafe_allow_html=True)
+        if st.button('💬 Chat'):
+            st.write('Reasoning with your inputs...')
+            user_prompt_sections = divide_prompt(user_prompt, max_length)
+            full_response = ''
+            for prompt_section in user_prompt_sections:
+                response = chat_with_model(prompt_section, ''.join(list(document_sections)), model_choice)
+                full_response += response + '\n'  # Combine the responses
+            response = full_response
+            st.write('Response:')
+            st.write(response)
+            filename = generate_filename(user_prompt, choice)
+            create_file(filename, user_prompt, response, should_save)
+            st.sidebar.markdown(get_table_download_link(filename), unsafe_allow_html=True)
+
+    # Compose a file sidebar of markdown md files:
+    all_files = glob.glob("*.md")
+    all_files = [file for file in all_files if len(os.path.splitext(file)[0]) >= 10]  # exclude files with short names
     all_files.sort(key=lambda x: (os.path.splitext(x)[1], x), reverse=True)  # sort by file type and file name in descending order
     if st.sidebar.button("🗑 Delete All"):
         for file in all_files:
@@ -762,36 +755,77 @@ def main():
             
             st.experimental_rerun()
 
+
+    # Function to encode file to base64
+    def get_base64_encoded_file(file_path):
+        with open(file_path, "rb") as file:
+            return base64.b64encode(file.read()).decode()
+
+    # Function to create a download link
+    def get_audio_download_link(file_path):
+        base64_file = get_base64_encoded_file(file_path)
+        return f'<a href="data:file/wav;base64,{base64_file}" download="{os.path.basename(file_path)}">⬇️ Download Audio</a>'
+
+    # Compose a file sidebar of past encounters
+    all_files = glob.glob("*.wav")
+    all_files = [file for file in all_files if len(os.path.splitext(file)[0]) >= 10]  # exclude files with short names
+    all_files.sort(key=lambda x: (os.path.splitext(x)[1], x), reverse=True)  # sort by file type and file name in descending order
+
+    filekey = 'del' + str(file)
+    if st.sidebar.button("🗑 Delete All", key=filekey):
+        for file in all_files:
+            os.remove(file)
+        st.experimental_rerun()
+
+    for file in all_files:
+        col1, col2 = st.sidebar.columns([6, 1])  # adjust the ratio as needed
+        with col1:
+            st.markdown(file)
+            if st.button("🎵", key="play_" + file):  # play emoji button
+                audio_file = open(file, 'rb')
+                audio_bytes = audio_file.read()
+                st.audio(audio_bytes, format='audio/wav')
+                #st.markdown(get_audio_download_link(file), unsafe_allow_html=True)
+                #st.text_input(label="", value=file)
+        with col2:
+            if st.button("🗑", key="delete_" + file):
+                os.remove(file)
+                st.experimental_rerun()
+
+
+
     # Feedback
     # Step: Give User a Way to Upvote or Downvote
-    feedback = st.radio("Step 8: Give your feedback", ("👍 Upvote", "👎 Downvote"))
-    if feedback == "👍 Upvote":
-        st.write("You upvoted 👍. Thank you for your feedback!")
-    else:
-        st.write("You downvoted 👎. Thank you for your feedback!")
-        
-    load_dotenv()
-    st.write(css, unsafe_allow_html=True)
-    st.header("Chat with documents :books:")
-    user_question = st.text_input("Ask a question about your documents:")
-    if user_question:
-        process_user_input(user_question)
-    with st.sidebar:
-        st.subheader("Your documents")
-        docs = st.file_uploader("import documents", accept_multiple_files=True)
-        with st.spinner("Processing"):
-            raw = pdf2txt(docs)
-            if len(raw) > 0:
-                length = str(len(raw))
-                text_chunks = txt2chunks(raw)
-                vectorstore = vector_store(text_chunks)
-                st.session_state.conversation = get_chain(vectorstore)
-                st.markdown('# AI Search Index of Length:' + length + ' Created.')  # add timing
-                filename = generate_filename(raw, 'txt')
-                create_file(filename, raw, '', should_save)
+    with st.expander("Give your feedback 👍", expanded=False):
+
+        feedback = st.radio("Step 8: Give your feedback", ("👍 Upvote", "👎 Downvote"))
+        if feedback == "👍 Upvote":
+            st.write("You upvoted 👍. Thank you for your feedback!")
+        else:
+            st.write("You downvoted 👎. Thank you for your feedback!")
+            
+        load_dotenv()
+        st.write(css, unsafe_allow_html=True)
+        st.header("Chat with documents :books:")
+        user_question = st.text_input("Ask a question about your documents:")
+        if user_question:
+            process_user_input(user_question)
+        with st.sidebar:
+            st.subheader("Your documents")
+            docs = st.file_uploader("import documents", accept_multiple_files=True)
+            with st.spinner("Processing"):
+                raw = pdf2txt(docs)
+                if len(raw) > 0:
+                    length = str(len(raw))
+                    text_chunks = txt2chunks(raw)
+                    vectorstore = vector_store(text_chunks)
+                    st.session_state.conversation = get_chain(vectorstore)
+                    st.markdown('# AI Search Index of Length:' + length + ' Created.')  # add timing
+                    filename = generate_filename(raw, 'txt')
+                    create_file(filename, raw, '', should_save)
 
 # 18. Run AI Pipeline
 if __name__ == "__main__":
     whisper_main()
     main()
-    add_Med_Licensing_Exam_Dataset()
+    #add_Med_Licensing_Exam_Dataset()