Update app.py

app.py

@@ -1,61 +1,116 @@
+import streamlit as st
 import requests
 import os
+from pytube import YouTube
 import tempfile
 from fpdf import FPDF
-import streamlit as st
-from pytube import YouTube
+import time
 
-os.environ['TOGETHER_API_KEY'] = 'ef5312e1d4bffafb5daf8ee0d632feb1aa94a021d020f4d0e6f6ded5b0795cc4'
+# Together.ai API configuration
+os.environ['TOGETHER_API_KEY'] = os.environ.get('TOGETHER_API_KEY')
 together_api_key = os.environ.get("TOGETHER_API_KEY")
-headers = {
-    "Authorization": f"Bearer {together_api_key}",
-    "Content-Type": "application/json"
+together_url = "https://api.together.xyz/inference"
+
+# AssemblyAI API configuration
+assembly_base_url = "https://api.assemblyai.com/v2"
+assembly_headers = {
+    "authorization": os.environ.get('ASSEMBLYAI_API_KEY')
 }
 
 def generate_sentiment_score(input_text, parameters):
     prompt = f'''
     You are an experienced interview reviewer and consultant for a reputable company. Your role is to evaluate the sentiment displayed by job candidates during their interviews based on the transcripts of their responses.
-    The parameters to evaluate are: {', '.join(parameters)}.
-    Provide a sentiment score on a scale of 1-5 for each parameter.
+    The hiring team has provided you with an interview transcript and has asked you to analyze the candidate's sentiment for the following parameters: {', '.join(parameters)}. Your assessment will help the team make more informed hiring decisions and identify candidates who demonstrate genuine positive sentiment towards the role and the company.
+    The parameters to evaluate are:
+    {', '.join(parameters)}.
+    To complete this task, you will:
+    1. Carefully review the provided interview transcript.
+    2. Consider phrases, word choices, or patterns of speech that convey positive or negative sentiment for each parameter.
+    3. Based on your analysis, provide a sentiment score on a scale of 1-5 for each parameter, with 1 being extremely negative and 5 being extremely positive.
+    Provide your scores in the format: Parameter: Score.
     '''
+
+    headers = {
+        "Authorization": f"Bearer {together_api_key}",
+        "Content-Type": "application/json"
+    }
+
     data = {
-        "model": "mistralai/Mixtral-8x7B-Instruct-v0.1",
-        "prompt": prompt,
+        "model": "togethercomputer/llama-2-70b-chat",
+        "prompt": f"[INST]{prompt}\n\n{input_text}[/INST]",
         "temperature": 0.0,
         "max_tokens": 1024
     }
-    response = requests.post("https://api.together.xyz/inference", headers=headers, json=data)
+
+    response = requests.post(together_url, headers=headers, json=data)
+
     if response.status_code == 200:
         return response.json()['output']['choices'][0]['text']
     else:
-        print(f"Error: {response.status_code} - {response.text}")
+        raise Exception(f"Error: {response.status_code} - {response.text}")
 
 def generate_detailed_feedback(input_text, parameters):
     prompt = f'''
-    Provide a detailed analysis of the candidate's responses based on the following parameters: {', '.join(parameters)}.
-    Include specific examples, quotes, and adjectives from the transcript that support your analysis. Summarize the candidate's overall sentiment and demeanor.
+    As an experienced interview reviewer, provide a detailed analysis of the candidate's responses based on the following parameters: {', '.join(parameters)}. 
+    
+    Include specific examples, quotes, and adjectives from the transcript that support your analysis. Offer actionable insights and recommendations for the hiring team to make informed decisions. Summarize the candidate's overall sentiment and demeanor.
     '''
+
+    headers = {
+        "Authorization": f"Bearer {together_api_key}",
+        "Content-Type": "application/json"
+    }
+
     data = {
-        "model": "mistralai/Mixtral-8x7B-Instruct-v0.1",
-        "prompt": prompt,
+        "model": "togethercomputer/llama-2-70b-chat",
+        "prompt": f"[INST]{prompt}\n\n{input_text}[/INST]",
         "temperature": 0.0,
-        "max_tokens": 1024
+        "max_tokens": 2048
     }
-    response = requests.post("https://api.together.xyz/inference", headers=headers, json=data)
+
+    response = requests.post(together_url, headers=headers, json=data)
+
     if response.status_code == 200:
         return response.json()['output']['choices'][0]['text']
     else:
-        print(f"Error: {response.status_code} - {response.text}")
+        raise Exception(f"Error: {response.status_code} - {response.text}")
+
+def upload_to_assemblyai(file_path):
+    with open(file_path, "rb") as f:
+        response = requests.post(f"{assembly_base_url}/upload", headers=assembly_headers, data=f)
+    return response.json().get("upload_url")
+
+def transcribe_with_assemblyai(upload_url):
+    data = {
+        "audio_url": upload_url
+    }
+    response = requests.post(f"{assembly_base_url}/transcript", json=data, headers=assembly_headers)
+    transcript_id = response.json().get("id")
+
+    while True:
+        response = requests.get(f"{assembly_base_url}/transcript/{transcript_id}", headers=assembly_headers)
+        result = response.json()
+        if result['status'] == 'completed':
+            return result['text']
+        elif result['status'] == 'failed':
+            raise Exception("Transcription failed")
+        time.sleep(5)
 
 def transcript(video_link):
     try:
         yt = YouTube(video_link)
         stream = yt.streams.filter(only_audio=True).first()
         temp_file_path = stream.download()
-        with open(temp_file_path, 'rb') as audio_data:
-            total_transcript = (openai.Audio.transcribe("whisper-1", audio_data))["text"]
+
+        print(f"Video '{yt.title}.mp4' downloaded successfully!")
+
+        upload_url = upload_to_assemblyai(temp_file_path)
+        transcription_text = transcribe_with_assemblyai(upload_url)
+
         os.remove(temp_file_path)
-        return total_transcript
+
+        return transcription_text
+
     except Exception as e:
         print(f"Error: {e}")
         return None
@@ -65,6 +120,31 @@ def main():
     st.title("Insight Hire")
     st.write("Analyze interview transcripts or videos to gain valuable insights into candidate sentiment.")
 
+    st.sidebar.markdown("## About")
+    st.sidebar.markdown("""
+    <div style='color: #1f77b4; font-weight: bold;'>Streamline Your Interview Evaluation</div>
+    - Get data-driven sentiment scores for key parameters
+    - Identify top candidates based on sentiment analysis
+    - Make informed hiring decisions with actionable insights
+    """, unsafe_allow_html=True)
+
+    st.sidebar.markdown("<hr>", unsafe_allow_html=True)  # Horizontal separator
+
+    st.sidebar.markdown("## Tips")
+    st.sidebar.markdown("""
+    <div style='color: #2ca02c; font-weight: bold;'>📝 Input Preparation</div>
+    - Provide clear interview transcripts or valid video links
+    - Specify relevant parameters for sentiment analysis
+    """, unsafe_allow_html=True)
+
+    st.sidebar.markdown("<hr>", unsafe_allow_html=True)  # Horizontal separator
+
+    st.sidebar.markdown("## About Me")
+    st.sidebar.markdown("""
+    <div style='color: #d62728; font-weight: bold;'>👋 Hi, I'm Dhruv!</div>
+    I want to make a real impact in the field of AI/ML . My main interest lies in model building and deployment. I'm passionate about leveraging cutting-edge technologies to solve real-world problems.
+    """, unsafe_allow_html=True)
+
     input_option = st.radio("Select input type", ("Text", "YouTube Video Link"))
 
     input_text = ""
@@ -87,7 +167,9 @@ def main():
             with st.spinner("Generating sentiment scores..."):
                 sentiment_scores = generate_sentiment_score(input_text, parameters)
                 sentiment_scores = sentiment_scores.strip().split("\n")
+
                 st.subheader("Sentiment Scores")
+                valid_scores = []
                 for score in sentiment_scores:
                     if ":" in score:
                         param, score_value = score.split(":")
@@ -96,22 +178,25 @@ def main():
                         if param in parameters:
                             try:
                                 score_value = float(score_value.split("/")[0].strip())
+                                valid_scores.append((param, score_value))
                                 if score_value >= 4:
-                                    color = "#2ca02c"
+                                    color = "#2ca02c"  # Green
                                 elif score_value >= 3:
-                                    color = "#ff7f0e"
+                                    color = "#ff7f0e"  # Orange
                                 else:
-                                    color = "#d62728"
+                                    color = "#d62728"  # Red
                                 st.markdown(f"**{param}**: <span style='color: {color}'>{score_value}/5</span>", unsafe_allow_html=True)
                             except ValueError:
                                 pass
 
             if valid_scores:
+                # Generate detailed feedback
                 with st.spinner("Generating detailed feedback..."):
                     detailed_feedback = generate_detailed_feedback(input_text, parameters)
                     st.subheader("Detailed Feedback")
                     st.write(detailed_feedback)
 
+                    # Provide an option to download detailed feedback as a .txt or .pdf file
                     temp_txt_path = tempfile.mktemp(suffix=".txt")
                     with open(temp_txt_path, 'w') as f:
                         f.write(detailed_feedback)
@@ -143,7 +228,7 @@ def main():
 
 if __name__ == "__main__":
     main()
-
+