initial commit

app.py

@@ -0,0 +1,55 @@
+import gradio as gr
+from dotenv import load_dotenv
+from gradio import ChatMessage
+from setup import  Speech_Text
+from temp import Script
+load_dotenv()
+
+transcriptor = Speech_Text()
+output_id = None
+database_con = Script()
+
+# Function to generate chatbot response
+def generate_response(chat_history: list[ChatMessage]):
+    return database_con.request(chat_history)
+
+def process(audio, input_text, chat_history: list[ChatMessage]):
+
+    if audio is not None:
+        transcript = transcriptor.get_transcript(audio)
+        chat_history.append({"role": "user", "content": transcript})
+
+    elif input_text:
+        chat_history.append({"role": "user", "content": input_text})
+
+    else:
+        response = 'Provide a query text or an audio to query.'
+        chat_history.append({"role": "assistant", "content": response})
+        audio_data = transcriptor.speech_synthesis(response)
+        return audio_data, chat_history
+
+    response = generate_response(chat_history)
+    chat_history.append({"role": "assistant", "content": response})
+    # audio_data = transcriptor.speech_synthesis(response)
+    return chat_history
+
+# Create Gradio Blocks interface
+with gr.Blocks() as demo:
+    
+    with gr.Row():
+        chatbot = gr.Chatbot(label="Chatbot Conversation", type="messages", bubble_full_width=True, show_copy_button=True, autoscroll=True)
+    
+    with gr.Row():
+        input_textbox = gr.Textbox(label="Input Text", placeholder="Type your message here...")
+        input_audio = gr.Audio(label="Input Audio", sources="microphone", type="numpy")
+    
+    process_button = gr.Button("Submit Query")
+
+    process_button.click(
+        fn=process,
+        inputs=[input_audio, input_textbox, chatbot],
+        outputs=[chatbot]
+    )
+
+if __name__ == "__main__":
+    demo.launch(server_name="0.0.0.0",server_port=80)

setup.py

@@ -0,0 +1,31 @@
+
+import os
+import io
+from groq import Groq
+import soundfile as sf
+from deepgram import DeepgramClient, SpeakOptions
+from langchain_groq import ChatGroq
+from dotenv import load_dotenv
+
+load_dotenv('.env')
+
+
+# Text to Speech and Speech to Text
+class Speech_Text():
+    def __init__(self):
+        self.client = Groq(api_key=os.getenv("GROQ_API_KEY"))
+        
+    # Function to get transcript from audio
+    def get_transcript(self,audio):
+        audio_buffer = io.BytesIO()
+        sf.write(audio_buffer, audio[1], samplerate=audio[0], format="MP3")
+        audio_buffer.seek(0)
+        translation = self.client.audio.transcriptions.create(
+            file=("audio.mp3", audio_buffer.read()),
+            model="distil-whisper-large-v3-en",
+            response_format="json",
+            temperature=0.0,
+        )
+        return translation.text
+
+    

temp.py

@@ -0,0 +1,41 @@
+from langchain_community.utilities import SerpAPIWrapper
+from langgraph.prebuilt import create_react_agent
+import getpass
+import os
+import sqlite3
+from langchain_openai import ChatOpenAI
+from langchain_community.tools.tavily_search import TavilySearchResults
+from langchain_core.messages import HumanMessage,AIMessage
+from langchain.agents import initialize_agent
+from langchain.agents.agent_types import AgentType
+from langchain_community.agent_toolkits.sql.toolkit import SQLDatabaseToolkit
+from langchain_community.utilities.sql_database import SQLDatabase
+from langchain_openai import ChatOpenAI
+from langchain_groq import ChatGroq
+
+from dotenv import load_dotenv
+load_dotenv('.env')
+
+class Script:
+    def __init__(self):
+        self.model = ChatOpenAI(model="gpt-4o-mini")
+        self.groq = ChatGroq(model='llama3-70b-8192')
+        self.db = SQLDatabase.from_uri("sqlite:///sample_database.db")
+        self.toolkit = SQLDatabaseToolkit(db=self.db, llm=self.model)
+        search = TavilySearchResults(max_results=2)
+        tools = self.toolkit.get_tools()+[search]
+        self.agent = create_react_agent(self.model, tools)
+
+    def history(self,hist):
+        message = []
+        for i in hist[-5:]:
+            if(i['role']=='user'):
+                message+=[HumanMessage(content=i['content'])]
+            else:
+                message+=[AIMessage(content=i['content'])]
+        return message
+
+    def request(self,message):
+        message = self.history(message)
+        response = self.agent.invoke({"messages": message})
+        return response["messages"][-1].content