Update app.py

app.py

@@ -4,14 +4,22 @@ import re
 from dotenv import load_dotenv
 import os
 from groq import Groq
+from openai import OpenAI
+import tempfile
+from pydub import AudioSegment
+import io
 
-# Load the API key from .env file
+# Load the API keys from .env file
 load_dotenv()
-api_key = os.getenv('GROQ_API_KEY')
-client = Groq(api_key=api_key)
+groq_api_key = os.getenv('GROQ_API_KEY')
+openai_api_key = os.getenv('OPENAI_API_KEY')
 
-if not api_key:
-    raise ValueError("API key is not set. Please check your .env file and ensure GROQ_API_KEY is set.")
+# Initialize clients
+groq_client = Groq(api_key=groq_api_key)
+openai_client = OpenAI(api_key=openai_api_key)
+
+if not groq_api_key or not openai_api_key:
+    raise ValueError("API keys are not set. Please check your .env file and ensure GROQ_API_KEY and OPENAI_API_KEY are set.")
 
 def get_transcript(url):
     try:
@@ -26,17 +34,41 @@ def get_transcript(url):
     except Exception as e:
         return f"Error: {str(e)}"
 
+def convert_to_supported_format(file):
+    audio = AudioSegment.from_file(file)
+    buffer = io.BytesIO()
+    audio.export(buffer, format="mp3")
+    buffer.seek(0)
+    return buffer
+
+def transcribe_audio(file):
+    file = convert_to_supported_format(file)
+    with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as temp_file:
+        temp_file.write(file.getvalue())
+        temp_file_path = temp_file.name
+
+    try:
+        with open(temp_file_path, "rb") as audio_file:
+            transcript = openai_client.audio.transcriptions.create(
+                model="whisper-1", 
+                file=audio_file
+            )
+        return transcript.text
+    except Exception as e:
+        return f"Error in transcription: {str(e)}"
+    finally:
+        os.remove(temp_file_path)
+
 def answer_question(transcript, question):
     try:
-        response = client.chat.completions.create(
+        response = groq_client.chat.completions.create(
             model="mixtral-8x7b-32768",
             messages=[
                 {"role": "system", "content": "You are a helpful assistant."},
-                {"role": "user", "content": f"Using the following transcript as context and then go through the question and then please solve or answer the question correctly in 300 words:\n\nTranscript:\n{transcript}\n\nQuestion:\n{question}"}
+                {"role": "user", "content": f"Using the following transcript as context, please answer the question:\n\nTranscript:\n{transcript}\n\nQuestion:\n{question}"}
             ],
-            max_tokens=500
+            max_tokens=150
         )
-
         if response.choices and response.choices[0].message:
             answer = response.choices[0].message.content.strip()
             return answer
@@ -45,24 +77,41 @@ def answer_question(transcript, question):
     except Exception as e:
         return f"Error in answering question: {str(e)}"
 
-def handle_query(youtube_url, question):
-    transcript = get_transcript(youtube_url)
+def handle_query(transcript, question):
     if "Error" in transcript:
         return transcript
     answer = answer_question(transcript, question)
     return answer
 
-st.title("YouTube Video Doubt Bot")
-youtube_url = st.text_input("YouTube URL", placeholder="Enter YouTube URL here...")
+st.title("Video/Audio Doubt Bot")
 
-if youtube_url:
-    transcript = get_transcript(youtube_url)
-    if "Error" in transcript:
-        st.write(transcript)
-    else:
-        st.write("Transcript successfully loaded.")
-        question = st.text_input("Ask a question about the video")
-        if st.button("Get Answer"):
-            answer = handle_query(youtube_url, question)
-            st.write("### Answer")
-            st.write(answer)
+option = st.selectbox("Choose input type", ("YouTube URL", "Upload audio/video file"))
+
+if option == "YouTube URL":
+    youtube_url = st.text_input("YouTube URL", placeholder="Enter YouTube URL here...")
+    if youtube_url:
+        transcript = get_transcript(youtube_url)
+        if "Error" in transcript:
+            st.write(transcript)
+        else:
+            st.write("Transcript successfully loaded.")
+            question = st.text_input("Ask a question about the video")
+            if st.button("Get Answer"):
+                answer = handle_query(transcript, question)
+                st.write("### Answer")
+                st.write(answer)
+
+elif option == "Upload audio/video file":
+    uploaded_file = st.file_uploader("Choose an audio or video file", type=["mp3", "mp4", "wav"])
+    if uploaded_file is not None:
+        with st.spinner('Transcribing audio...'):
+            transcript = transcribe_audio(uploaded_file)
+        if "Error" in transcript:
+            st.write(transcript)
+        else:
+            st.write("File successfully transcribed.")
+            question = st.text_input("Ask a question about the audio/video")
+            if st.button("Get Answer"):
+                answer = handle_query(transcript, question)
+                st.write("### Answer")
+                st.write(answer)