Update app.py

app.py

@@ -1,71 +1,52 @@
 import streamlit as st
-from tempfile import NamedTemporaryFile
-from audiorecorder import audiorecorder
-from whispercpp import Whisper
-from pydub import AudioSegment
-import io
+from audio_recorder_streamlit import audio_recorder
+from openai import OpenAI
+API_KEY = 'enter-openai-api-key-here'
 
-# Initialize whisper.cpp
-w = Whisper('tiny')
 
-def inference(audio_segment):
-    # Convert AudioSegment to WAV format in memory
-    with NamedTemporaryFile(suffix=".wav", delete=False) as temp:
-        # Export AudioSegment to raw bytes in WAV format
-        audio_segment.export(temp.name, format="wav")
-        temp.close()  # Ensure the file is written and closed before passing it to Whisper
-        result = w.transcribe(temp.name)
-        text = w.extract_text(result)
-    return text[0]
 
-# Streamlit UI setup
-with st.sidebar:
-    audio = audiorecorder("Click to send voice message", "Recording... Click when you're done", key="recorder")
-    st.title("Echo Bot with Whisper")
+def transcribe_text_to_voice(audio_location):
+    client = OpenAI(api_key=API_KEY)
+    audio_file= open(audio_location, "rb")
+    transcript = client.audio.transcriptions.create(model="whisper-1", file=audio_file)
+    return transcript.text
 
-# Initialize chat history
-if "messages" not in st.session_state:
-    st.session_state.messages = []
+def chat_completion_call(text):
+    client = OpenAI(api_key=API_KEY)
+    messages = [{"role": "user", "content": text}]
+    response = client.chat.completions.create(model="gpt-3.5-turbo-1106", messages=messages)
+    return response.choices[0].message.content
 
-# Display chat messages from history on app rerun
-for message in st.session_state.messages:
-    with st.chat_message(message["role"]):
-        st.markdown(message["content"])
 
-# React to user input
-if (prompt := st.chat_input("Your message")) or len(audio):
-    # If it's coming from the audio recorder transcribe the message with whisper.cpp
-    if len(audio) > 0:
-        # Debugging: Check the type of the audio object
-        st.write(f"Audio Type: {type(audio)}")
+def text_to_speech_ai(speech_file_path, api_response):
+    client = OpenAI(api_key=API_KEY)
+    response = client.audio.speech.create(model="tts-1",voice="nova",input=api_response)
+    response.stream_to_file(speech_file_path)
 
-        # Handle the case where audio is in a byte format
-        if isinstance(audio, bytes):
-            try:
-                # Convert the raw byte data to an AudioSegment instance
-                audio_segment = AudioSegment.from_file(io.BytesIO(audio), format="wav")
-                prompt = inference(audio_segment)
-            except Exception as e:
-                st.error(f"Error processing audio: {e}")
-                prompt = "Sorry, there was an error processing your audio."
 
-        # Handle the case where audio is an AudioSegment object
-        elif isinstance(audio, AudioSegment):
-            # Process it directly since it's already an AudioSegment
-            prompt = inference(audio)
 
-        else:
-            st.error("The audio data is not in the expected format.")
-            prompt = "Sorry, the audio format is not correct."
+st.title("🧑‍💻 Skolo Online 💬 Talking Assistant")
 
-    # Display user message in chat message container
-    st.chat_message("user").markdown(prompt)
-    # Add user message to chat history
-    st.session_state.messages.append({"role": "user", "content": prompt})
+"""
+Hi🤖 just click on the voice recorder and let me know how I can help you today?
+"""
 
-    response = f"Echo: {prompt}"
-    # Display assistant response in chat message container
-    with st.chat_message("assistant"):
-        st.markdown(response)
-    # Add assistant response to chat history
-    st.session_state.messages.append({"role": "assistant", "content": response})
+audio_bytes = audio_recorder()
+if audio_bytes:
+    ##Save the Recorded File
+    audio_location = "audio_file.wav"
+    with open(audio_location, "wb") as f:
+        f.write(audio_bytes)
+
+    #Transcribe the saved file to text
+    text = transcribe_text_to_voice(audio_location)
+    st.write(text)
+
+    #Use API to get an AI response
+    api_response = chat_completion_call(text)
+    st.write(api_response)
+
+    # Read out the text response using tts
+    speech_file_path = 'audio_response.mp3'
+    text_to_speech_ai(speech_file_path, api_response)
+    st.audio(speech_file_path)