Upload 4 files

.env

@@ -0,0 +1 @@
+API_KEY='AIzaSyBeNMjIT8iPUcnO2zzu7hWkfJFRxFwQTUg'

app.py

@@ -0,0 +1,51 @@
+import streamlit as st
+import time
+import sounddevice as sd
+from scipy.io.wavfile import write
+import numpy as np
+import soundfile as sf
+from utils.functions import voice_to_text, get_gemini_response, text_to_audio
+import pygame
+
+def main():
+    st.title("Mental Health Chatbot")
+
+    if "messages" not in st.session_state:
+        st.session_state.messages = []
+
+    for message in st.session_state.messages:
+        with st.container():
+            if message["role"] == "user":
+                st.markdown(f":speech_balloon: **You:** {message['content']}")
+            else:
+                st.markdown(f":robot: **Chatbot:** {message['content']}")
+
+    if st.button("Speak"):
+        with st.spinner('Recording...'):
+            try:
+                fs = 44100
+                seconds = 5
+                myrecording = sd.rec(int(seconds * fs), samplerate=fs, channels=1)
+                sd.wait()
+                audio_file = "user_audio.wav"
+                write(audio_file, fs, myrecording)
+
+                text_input = voice_to_text(audio_file)
+                if text_input != "Error during transcription":
+                    st.session_state.messages.append({"role": "user", "content": text_input})
+                    response = get_gemini_response(text_input)
+                    if response:
+                        st.session_state.messages.append({"role": "assistant", "content": response})
+                        audio_array = text_to_audio(response)
+                        bot_audio_file = "bot_audio.wav"
+                        sf.write(bot_audio_file, audio_array, 44100)
+                        pygame.mixer.init()
+                        pygame.mixer.music.load(bot_audio_file)
+                        pygame.mixer.music.play()
+                        while pygame.mixer.music.get_busy():
+                            time.sleep(1)
+            except Exception as e:
+                st.error(f"Error: {e}")
+
+if __name__ == "__main__":
+    main()

requirements.txt

@@ -0,0 +1,11 @@
+streamlit
+transformers
+sounddevice
+soundfile
+pygame
+torch
+numpy
+scipy
+requests
+google.generativeai
+time

utils/functions.py

@@ -0,0 +1,39 @@
+import torch
+from transformers import pipeline
+import requests
+import json
+import soundfile as sf
+import os
+import google.generativeai as genai
+
+
+def voice_to_text(audio_data, model_name="openai/whisper-small", device="cuda" if torch.cuda.is_available() else "cpu"):
+
+  try:
+    model = pipeline("automatic-speech-recognition", model=model_name, device=device, trust_remote_code=True)
+    text = model(audio_data)["text"]
+    return text
+  
+  except Exception as e:
+    return f"Error during transcription"
+
+
+def get_gemini_response(text):
+
+  genai.configure(api_key=os.environ['API_KEY'])
+  model = genai.GenerativeModel('gemini-1.5-flash')
+
+  prompt = f"You are a compassionate and supportive mental health assistant. Provide helpful advice, encouragement, and information to the user. Respond in a warm and understanding tone. User: {text}"
+
+  response = model.generate_content(prompt)
+  return response.text
+
+
+
+def text_to_audio(text):
+
+  device = "cuda" if torch.cuda.is_available() else "cpu"
+  model = pipeline("text-to-speech", model="parler-tts/parler_tts_mini_v0.1", device=device, trust_remote_code=True)
+
+  audio_array = model(text)["audio"]
+  return audio_array