Update app.py

app.py

@@ -1,36 +1,32 @@
 
 import gradio as gr
-from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline, WhisperProcessor, WhisperForConditionalGeneration
+from transformers import (
+    AutoTokenizer,
+    AutoModelForCausalLM,
+    pipeline,
+    WhisperProcessor,
+    WhisperForConditionalGeneration,
+)
 from langchain_community.document_loaders import PyPDFLoader
 from langchain.text_splitter import RecursiveCharacterTextSplitter
-from langchain.embeddings import HuggingFaceEmbeddings
-from langchain.vectorstores import Chroma
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain_community.vectorstores import Chroma
 from langchain.chains import RetrievalQA
-from langchain.llms import HuggingFacePipeline
+from langchain_community.llms import HuggingFacePipeline
 import torch
 import os
 
-
 # --------------------------
 # Load Pretrained LLaMA 2 via Hugging Face Hub
 # --------------------------
 llama_model_id = "meta-llama/Llama-2-7b-chat-hf"  # Replace with your model ID if different
 
-from huggingface_hub import login
-from transformers import AutoTokenizer
-
-
-from huggingface_hub import login
-
-hf_token = os.environ.get("HF_TOKEN")
-if hf_token:
-    login(token=hf_token)
-else:
-    raise ValueError("HF_TOKEN environment variable not found")
-
-tokenizer = AutoTokenizer.from_pretrained(model_id)
-
+hf_token = os.getenv("HF_TOKEN")
+if hf_token is None:
+    raise ValueError("HF_TOKEN environment variable not found. Please add it in your secrets or environment.")
 
+tokenizer = AutoTokenizer.from_pretrained(llama_model_id, token=hf_token)
+model = AutoModelForCausalLM.from_pretrained(llama_model_id, token=hf_token)
 
 # Text generation pipeline
 text_pipe = pipeline(
@@ -39,7 +35,7 @@ text_pipe = pipeline(
     tokenizer=tokenizer,
     max_new_tokens=256,
     temperature=0.7,
-    top_p=0.9
+    top_p=0.9,
 )
 
 # LangChain LLM wrapper
@@ -62,8 +58,8 @@ whisper_processor = WhisperProcessor.from_pretrained("openai/whisper-small")
 whisper_model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-small")
 
 def transcribe_audio(audio):
-    audio = whisper_processor(audio["array"], sampling_rate=audio["sampling_rate"], return_tensors="pt")
-    result = whisper_model.generate(**audio)
+    audio_input = whisper_processor(audio["array"], sampling_rate=audio["sampling_rate"], return_tensors="pt")
+    result = whisper_model.generate(**audio_input)
     return whisper_processor.batch_decode(result, skip_special_tokens=True)[0]
 
 # --------------------------