Update app.py

app.py

@@ -1,22 +1,39 @@
+import os
 import streamlit as st
 from transformers import pipeline, AutoProcessor, AutoModelForSpeechSeq2Seq, AutoTokenizer, AutoModelForCausalLM
 from espnet2.bin.tts_inference import Text2Speech
 import soundfile as sf
 from pydub import AudioSegment
 import io
+from dotenv import load_dotenv
+
+# Load environment variables from .env file
+load_dotenv()
+
+# Hugging Face token
+HUGGINGFACE_TOKEN = os.getenv("HUGGINGFACE_TOKEN")
+if not HUGGINGFACE_TOKEN:
+    st.error("Hugging Face token not found in .env file. Please add it.")
+    st.stop()
 
 # Load models
 @st.cache(allow_output_mutation=True)
 def load_models():
     # Speech-to-Text
-    processor = AutoProcessor.from_pretrained("openai/whisper-small")
-    stt_model = AutoModelForSpeechSeq2Seq.from_pretrained("openai/whisper-small")
-    stt_pipe = pipeline("automatic-speech-recognition", model=stt_model, tokenizer=processor.tokenizer, feature_extractor=processor.feature_extractor)
+    processor = AutoProcessor.from_pretrained("openai/whisper-small", use_auth_token=HUGGINGFACE_TOKEN)
+    stt_model = AutoModelForSpeechSeq2Seq.from_pretrained("openai/whisper-small", use_auth_token=HUGGINGFACE_TOKEN)
+    stt_pipe = pipeline(
+        "automatic-speech-recognition",
+        model=stt_model,
+        tokenizer=processor.tokenizer,
+        feature_extractor=processor.feature_extractor,
+        use_auth_token=HUGGINGFACE_TOKEN
+    )
 
     # Text Generation
-    tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.1")
-    text_model = AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.1")
-    text_pipe = pipeline("text-generation", model=text_model, tokenizer=tokenizer)
+    tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.1", use_auth_token=HUGGINGFACE_TOKEN)
+    text_model = AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.1", use_auth_token=HUGGINGFACE_TOKEN)
+    text_pipe = pipeline("text-generation", model=text_model, tokenizer=tokenizer, use_auth_token=HUGGINGFACE_TOKEN)
 
     # Text-to-Speech
     tts_model = Text2Speech.from_pretrained("espnet/espnet_tts_vctk_espnet_spk_voxceleb12_rawnet")
@@ -25,6 +42,7 @@ def load_models():
 
 stt_pipe, text_pipe, tts_model = load_models()
 
+# Streamlit app
 st.title("Voice-Enabled Chatbot")
 
 # Audio input