final

app.py

@@ -0,0 +1,131 @@
+import streamlit as st
+from whisper_stt import whisper_stt
+from st_pages import Page, show_pages
+from openai import OpenAI
+
+# Set page configuration
+st.set_page_config(layout="wide")
+show_pages([Page("app.py", "Home", "🏠")])
+
+# Initialize session state variables
+if 'paused' not in st.session_state:
+    st.session_state.paused = False
+if 'question_text' not in st.session_state:
+    st.session_state.question_text = ""
+if 'submitted' not in st.session_state:
+    st.session_state.submitted = False   
+if 'response_content' not in st.session_state:
+    st.session_state.response_content = ""
+if 'stopped' not in st.session_state:
+    st.session_state.stopped = False
+if 'function_call_count' not in st.session_state:
+    st.session_state.function_call_count = 0       
+if 'transcribed_text' not in st.session_state:
+    st.session_state.transcribed_text = ""
+if 'last_processed_text' not in st.session_state:
+    st.session_state.last_processed_text = ""
+if 'headers' not in st.session_state:
+    st.session_state.headers = []
+
+def create_anchor_link(text):
+    if text is None:
+        return ""
+    return f"<a href='#{text.strip().lower().replace(' ', '-').replace(',', '').replace('.', '').replace(chr(39), '-')}'>{text}</a>"
+
+def on_stop():
+    st.session_state.stopped = True
+
+col0 = st.columns(1)[0]
+placeholder = col0.empty()  # Define the placeholder
+def handle_enter(key):
+    if key == "ctrl+enter":
+        new_question = st.session_state.question_input
+        print(f"handle_enter called. new_question: '{new_question}'")
+        print(f"session state: {st.session_state}")
+with st.sidebar:
+    api_key = st.text_input("API Key", key="chatbot_api_key", type="password")
+    
+    col1, col2 = st.columns(2)
+
+    with col1:
+        # Call whisper_stt without a callback
+        transcribed_text = whisper_stt(
+            openai_api_key=api_key,
+            language='en'
+        )
+        if transcribed_text:
+            st.session_state.question_text = transcribed_text
+
+        # Check if new transcription is available
+        if transcribed_text and transcribed_text != st.session_state.transcribed_text:
+            st.session_state.transcribed_text = transcribed_text
+            st.session_state.question_text = transcribed_text
+            st.session_state.submitted = True
+
+    if st.session_state.question_text:
+        st.markdown(create_anchor_link(st.session_state.question_text), unsafe_allow_html=True)  
+
+    if 'question_input' in st.session_state and st.session_state.question_input:
+        st.markdown(create_anchor_link(st.session_state.question_input), unsafe_allow_html=True)         
+    with col2:
+        st.button(label='Stop', on_click=on_stop)
+   
+    # Create an input for the question and use new_question directly
+    new_question = st.text_area("Question", 
+        value=st.session_state.question_text or "", 
+        height=150, 
+        key="question_input",
+        on_change=handle_enter,
+        args=("ctrl+enter",)
+    )
+    print(f"After text_area, new_question: '{new_question}'")
+    # Check if new_question has changed and is not empty
+    if new_question and new_question != st.session_state.question_text:
+        st.session_state.question_text = new_question
+        st.session_state.submitted = True
+
+    st.markdown("## Navigation")
+    for header in st.session_state.headers:
+        st.markdown(create_anchor_link(header), unsafe_allow_html=True)    
+
+if st.session_state.question_text and not api_key:
+    st.info("Please add your OpenAI API key to continue.")
+    st.stop()
+
+if st.session_state.submitted and not st.session_state.stopped:
+    st.session_state.headers.append(st.session_state.question_text)
+    
+    if st.session_state.function_call_count == 0:
+        header = f'## {st.session_state.question_text}'
+        anchor = create_anchor_link(st.session_state.question_text)
+        st.session_state.response_content += f'{header}\n\n{anchor}\n\n'
+    else:
+        header = f'\n\n---\n{st.session_state.question_text}\n---\n\n'
+        anchor = create_anchor_link(st.session_state.question_text)
+        st.session_state.response_content += f'{header}{anchor}\n\n'
+    
+    st.session_state.function_call_count += 1
+    client = OpenAI(api_key=api_key)
+    st.session_state.messages = [{"role": "user", "content": st.session_state.question_text}]
+
+    response = client.chat.completions.create(
+        model="gpt-4o",
+        messages=st.session_state.messages,
+        stream=True
+    )
+
+    for chunk in response:
+        if st.session_state.stopped:
+            st.session_state.stopped = False
+            st.session_state.submitted = False
+            break
+        else:
+            if chunk and chunk.choices[0].delta.content:
+                st.session_state.response_content += chunk.choices[0].delta.content
+                placeholder.markdown(st.session_state.response_content, unsafe_allow_html=True)
+
+    st.session_state.submitted = False
+    st.session_state.stopped = False
+
+placeholder.markdown(st.session_state.response_content, unsafe_allow_html=True)
+st.session_state.stopped = False

push.bat

@@ -0,0 +1,3 @@
+git add .
+git commit -m "%1"
+git push -u origin 

requirements.txt

@@ -0,0 +1,94 @@
+aiohttp==3.9.5
+aiosignal==1.3.1
+altair==5.3.0
+annotated-types==0.7.0
+anthropic==0.31.2
+anyio==4.4.0
+attrs==23.2.0
+blinker==1.8.2
+bokeh==3.5.0
+cachetools==5.4.0
+certifi==2024.7.4
+charset-normalizer==3.3.2
+click==8.1.7
+colorama==0.4.6
+contourpy==1.2.1
+dataclasses-json==0.6.7
+distro==1.9.0
+duckduckgo_search==6.2.1
+filelock==3.15.4
+frozenlist==1.4.1
+fsspec==2024.6.1
+gitdb==4.0.11
+GitPython==3.1.43
+greenlet==3.0.3
+h11==0.14.0
+httpcore==1.0.5
+httpx==0.27.0
+huggingface-hub==0.23.5
+idna==3.7
+Jinja2==3.1.4
+jiter==0.5.0
+jsonpatch==1.33
+jsonpointer==3.0.0
+jsonschema==4.23.0
+jsonschema-specifications==2023.12.1
+langchain==0.2.9
+langchain-community==0.2.7
+langchain-core==0.2.21
+langchain-text-splitters==0.2.2
+langsmith==0.1.90
+loguru==0.7.2
+markdown-it-py==3.0.0
+MarkupSafe==2.1.5
+marshmallow==3.21.3
+mdurl==0.1.2
+multidict==6.0.5
+mypy-extensions==1.0.0
+numpy==1.26.4
+openai==1.35.14
+orjson==3.10.6
+packaging==24.1
+pandas==2.2.2
+pillow==10.4.0
+protobuf==5.27.2
+pyarrow==17.0.0
+pydantic==2.8.2
+pydantic_core==2.20.1
+pydeck==0.9.1
+Pygments==2.18.0
+pyreqwest_impersonate==0.5.0
+python-dateutil==2.9.0.post0
+pytz==2024.1
+PyYAML==6.0.1
+referencing==0.35.1
+requests==2.32.3
+rich==13.7.1
+rpds-py==0.19.0
+shellingham==1.5.4
+six==1.16.0
+smmap==5.0.1
+sniffio==1.3.1
+SpeechRecognition==3.10.4
+SQLAlchemy==2.0.31
+st-pages==0.5.0
+streamlit==1.36.0
+streamlit-bokeh-events==0.1.2
+streamlit-feedback==0.1.3
+streamlit_mic_recorder==0.0.8
+tenacity==8.5.0
+tokenizers==0.19.1
+toml==0.10.2
+toolz==0.12.1
+tornado==6.4.1
+tqdm==4.66.4
+trubrics==1.6.2
+typer==0.12.3
+typing-inspect==0.9.0
+typing_extensions==4.12.2
+tzdata==2024.1
+urllib3==2.2.2
+watchdog==4.0.1
+win32-setctime==1.1.0
+xyzservices==2024.6.0
+yarl==1.9.4

whisper_stt.py

@@ -0,0 +1,56 @@
+from streamlit_mic_recorder import mic_recorder
+import streamlit as st
+import io
+from openai import OpenAI
+import os
+
+
+def whisper_stt(openai_api_key=None, start_prompt="Start recording", stop_prompt="Stop recording", just_once=False,
+               use_container_width=False, language=None, callback=None, args=(), kwargs=None, key=None):
+    if not 'openai_client' in st.session_state:
+        st.session_state.openai_client = OpenAI(api_key=openai_api_key or os.getenv('OPENAI_API_KEY'))
+    if not '_last_speech_to_text_transcript_id' in st.session_state:
+        st.session_state._last_speech_to_text_transcript_id = 0
+    if not '_last_speech_to_text_transcript' in st.session_state:
+        st.session_state._last_speech_to_text_transcript = None
+    if key and not key + '_output' in st.session_state:
+        st.session_state[key + '_output'] = None
+    audio = mic_recorder(start_prompt=start_prompt, stop_prompt=stop_prompt, just_once=just_once,
+                         use_container_width=use_container_width,format="webm", key=key)
+    new_output = False
+    if audio is None:
+        output = None
+    else:
+        id = audio['id']
+        new_output = (id > st.session_state._last_speech_to_text_transcript_id)
+        if new_output:
+            output = None
+            st.session_state._last_speech_to_text_transcript_id = id
+            audio_bio = io.BytesIO(audio['bytes'])
+            audio_bio.name = 'audio.webm'
+            success = False
+            err = 0
+            while not success and err < 3:  # Retry up to 3 times in case of OpenAI server error.
+                try:
+                    transcript = st.session_state.openai_client.audio.transcriptions.create(
+                        model="whisper-1",
+                        file=audio_bio,
+                        language=language
+                    )
+                except Exception as e:
+                    print(str(e))  # log the exception in the terminal
+                    err += 1
+                else:
+                    success = True
+                    output = transcript.text
+                    st.session_state._last_speech_to_text_transcript = output
+        elif not just_once:
+            output = st.session_state._last_speech_to_text_transcript
+        else:
+            output = None
+
+    if key:
+        st.session_state[key + '_output'] = output
+    if new_output and callback:
+        callback(*args, **(kwargs or {}))
+    return output