Update process_interview.py

process_interview.py

@@ -355,11 +355,11 @@ def analyze_interviewee_voice(audio_path: str, utterances: List[Dict]) -> Dict:
         jitter = np.mean(np.abs(np.diff(pitches))) / pitch_mean if len(pitches) > 1 and pitch_mean > 0 else 0
         intensities = []
         for segment in segments:
-            rms = librosa.feature.rms(y=segment)[0]
+            rms = np.mean(librosa.feature.rms(y=segment)[0])
             intensities.extend(rms)
         intensity_mean = np.mean(intensities) if intensities else 0
         intensity_std = np.std(intensities) if intensities else 0
-        shimmer = np.mean(np.abs(np.diff(intensities))) / intensity_mean if len(intensities) > 1 and intensity_mean > 0 else 0
+        shimmer = np.mean(np.abs(np.diff(intensities))) / intensity_mean if intensity_mean > 0 else 0
         anxiety_score = 0.6 * (pitch_std / pitch_mean) + 0.4 * (jitter + shimmer) if pitch_mean > 0 else 0
         confidence_score = 0.7 * (1 / (1 + intensity_std)) + 0.3 * (1 / (1 + filler_ratio))
         hesitation_score = filler_ratio + repetition_score
@@ -446,7 +446,7 @@ def generate_report(analysis_data: Dict) -> str:
     try:
         voice = analysis_data.get('voice_analysis', {})
         voice_interpretation = generate_voice_interpretation(voice)
-        interviewee_responses = [f"- {u['text']}" for u in analysis_data['transcript'] if u['role'] == 'Interviewee'][:5]
+        interviewee_responses = [f"- {u['text']}" for u in analysis_data['transcript'] if u['role'] == 'Interviewee'][:3]
         acceptance_prob = analysis_data.get('acceptance_probability', 50.0)
         acceptance_line = f"\n**Suitability Score: {acceptance_prob:.2f}%**\n"
         if acceptance_prob >= 80:
@@ -458,30 +458,29 @@ def generate_report(analysis_data: Dict) -> str:
         else:
             acceptance_line += "HR Verdict: Limited fit, significant improvement required."
         prompt = f"""
-        You are EvalBot, a senior HR consultant delivering a concise, professional interview analysis report. Use clear headings, bullet points ('-'), and avoid redundancy. Ensure text is clean and free of special characters that could break formatting.
+        You are EvalBot, a senior HR consultant delivering a concise, professional interview analysis report. Use clear headings, bullet points ('-'), complete sentences, and formal language. Avoid redundancy, vague terms, and special characters that could break formatting. Ensure each section is unique and actionable.
         {acceptance_line}
         **1. Executive Summary**
-        - Summarize performance, key metrics, and hiring potential.
+        - Provide a narrative overview of the candidate’s performance, highlighting key strengths and fit.
         - Duration: {analysis_data['text_analysis']['total_duration']:.2f} seconds
         - Speaker Turns: {analysis_data['text_analysis']['speaker_turns']}
         - Participants: {', '.join(sorted(set(u['speaker'] for u in analysis_data['transcript'])))}
         **2. Communication and Vocal Dynamics**
-        - Evaluate vocal delivery (rate, fluency, confidence).
-        - Provide HR insights on workplace alignment.
+        - Evaluate vocal delivery (rate, fluency, confidence) with specific insights.
         {voice_interpretation}
         **3. Competency and Content**
-        - Assess leadership, problem-solving, communication, adaptability.
-        - List strengths and growth areas separately with examples.
+        - Assess leadership, problem-solving, communication, and adaptability with clear examples.
+        - List strengths and growth areas separately, using quantifiable achievements where possible.
         - Sample responses:
         {chr(10).join(interviewee_responses)}
         **4. Role Fit and Potential**
-        - Analyze cultural fit, role readiness, and growth potential.
+        - Analyze cultural fit, role readiness, and long-term growth potential with specific alignment to role requirements.
         **5. Recommendations**
-        - Provide prioritized strategies for growth (communication, technical skills, presence).
-        - Suggest next steps for hiring managers (advance, train, assess).
+        - Provide prioritized development strategies (e.g., communication training, technical assessments).
+        - Suggest specific next steps for hiring managers (e.g., advance, schedule tests).
         """
         response = gemini_model.generate_content(prompt)
-        return re.sub(r'[^\x00-\x7F]+', '', response.text)  # Sanitize non-ASCII characters
+        return re.sub(r'[^\x00-\x7F]+|[()]+', '', response.text)  # Sanitize non-ASCII and parentheses
     except Exception as e:
         logger.error(f"Report generation failed: {str(e)}")
         return f"Error generating report: {str(e)}"
@@ -492,36 +491,36 @@ def create_pdf_report(analysis_data: Dict, output_path: str, gemini_report_text:
                                 rightMargin=0.75*inch, leftMargin=0.75*inch,
                                 topMargin=1*inch, bottomMargin=1*inch)
         styles = getSampleStyleSheet()
-        h1 = ParagraphStyle(name='Heading1', fontSize=20, leading=24, spaceAfter=18, alignment=1, textColor=colors.HexColor('#003087'), fontName='Helvetica-Bold')
-        h2 = ParagraphStyle(name='Heading2', fontSize=14, leading=16, spaceBefore=12, spaceAfter=8, textColor=colors.HexColor('#0050BC'), fontName='Helvetica-Bold')
-        h3 = ParagraphStyle(name='Heading3', fontSize=10, leading=12, spaceBefore=8, spaceAfter=6, textColor=colors.HexColor('#3F7CFF'), fontName='Helvetica')
-        body_text = ParagraphStyle(name='BodyText', fontSize=9, leading=12, spaceAfter=6, fontName='Helvetica', textColor=colors.HexColor('#333333'))
-        bullet_style = ParagraphStyle(name='Bullet', parent=body_text, leftIndent=18, bulletIndent=8, fontName='Helvetica', bulletFontName='Helvetica', bulletFontSize=9)
+        h1 = ParagraphStyle(name='Heading1', fontSize=18, leading=22, spaceAfter=16, alignment=1, textColor=colors.HexColor('#003087'), fontName='Helvetica-Bold')
+        h2 = ParagraphStyle(name='Heading2', fontSize=13, leading=15, spaceBefore=10, spaceAfter=6, textColor=colors.HexColor('#0050BC'), fontName='Helvetica-Bold')
+        h3 = ParagraphStyle(name='Heading3', fontSize=9, leading=11, spaceBefore=6, spaceAfter=4, textColor=colors.HexColor('#3F7CFF'), fontName='Helvetica')
+        body_text = ParagraphStyle(name='BodyText', fontSize=8, leading=10, spaceAfter=4, fontName='Helvetica', textColor=colors.HexColor('#333333'))
+        bullet_style = ParagraphStyle(name='Bullet', parent=body_text, leftIndent=16, bulletIndent=6, fontName='Helvetica', bulletFontName='Helvetica', bulletFontSize=8)
         
         story = []
 
         def header_footer(canvas, doc):
             canvas.saveState()
-            canvas.setFont('Helvetica', 8)
+            canvas.setFont('Helvetica', 7)
             canvas.setFillColor(colors.HexColor('#666666'))
             canvas.drawString(doc.leftMargin, 0.5*inch, f"Page {doc.page} | EvalBot HR Interview Report | Confidential")
             canvas.setStrokeColor(colors.HexColor('#0050BC'))
-            canvas.setLineWidth(0.8)
+            canvas.setLineWidth(0.5)
             canvas.line(doc.leftMargin, doc.height + 0.9*inch, doc.width + doc.leftMargin, doc.height + 0.9*inch)
-            canvas.setFont('Helvetica-Bold', 9)
+            canvas.setFont('Helvetica-Bold', 8)
             canvas.drawString(doc.leftMargin, doc.height + 0.95*inch, "Candidate Interview Analysis")
             canvas.drawRightString(doc.width + doc.leftMargin, doc.height + 0.95*inch, time.strftime('%B %d, %Y'))
             canvas.restoreState()
 
         # Title Page
         story.append(Paragraph("Candidate Interview Analysis", h1))
-        story.append(Paragraph(f"Generated: {time.strftime('%B %d, %Y')}", ParagraphStyle(name='Date', alignment=1, fontSize=9, textColor=colors.HexColor('#666666'), fontName='Helvetica')))
-        story.append(Spacer(1, 0.4*inch))
+        story.append(Paragraph(f"Generated: {time.strftime('%B %d, %Y')}", ParagraphStyle(name='Date', alignment=1, fontSize=8, textColor=colors.HexColor('#666666'), fontName='Helvetica')))
+        story.append(Spacer(1, 0.3*inch))
         acceptance_prob = analysis_data.get('acceptance_probability', 50.0)
         story.append(Paragraph("Hiring Suitability Snapshot", h2))
         prob_color = colors.HexColor('#2E7D32') if acceptance_prob >= 80 else (colors.HexColor('#F57C00') if acceptance_prob >= 60 else colors.HexColor('#D32F2F'))
-        story.append(Paragraph(f"Suitability Score: <font size=15 color='{prob_color.hexval()}'><b>{acceptance_prob:.2f}%</b></font>",
-                             ParagraphStyle(name='Prob', fontSize=11, spaceAfter=10, alignment=1, fontName='Helvetica-Bold')))
+        story.append(Paragraph(f"Suitability Score: <font size=14 color='{prob_color.hexval()}'><b>{acceptance_prob:.2f}%</b></font>",
+                             ParagraphStyle(name='Prob', fontSize=10, spaceAfter=8, alignment=1, fontName='Helvetica-Bold')))
         if acceptance_prob >= 80:
             story.append(Paragraph("<b>HR Verdict:</b> Outstanding candidate, recommended for immediate advancement.", body_text))
         elif acceptance_prob >= 60:
@@ -530,28 +529,29 @@ def create_pdf_report(analysis_data: Dict, output_path: str, gemini_report_text:
             story.append(Paragraph("<b>HR Verdict:</b> Moderate potential, needs additional assessment.", body_text))
         else:
             story.append(Paragraph("<b>HR Verdict:</b> Limited fit, significant improvement required.", body_text))
-        story.append(Spacer(1, 0.3*inch))
+        story.append(Spacer(1, 0.2*inch))
+        participants = sorted([p for p in set(u['speaker'] for u in analysis_data['transcript']) if p != 'Unknown'])
         table_data = [
             ['Metric', 'Value'],
             ['Interview Duration', f"{analysis_data['text_analysis']['total_duration']:.2f} seconds"],
             ['Speaker Turns', f"{analysis_data['text_analysis']['speaker_turns']}"],
-            ['Participants', ', '.join(sorted(set(u['speaker'] for u in analysis_data['transcript'])))],
+            ['Participants', ', '.join(participants)],
         ]
-        table = Table(table_data, colWidths=[2.3*inch, 3.7*inch])
+        table = Table(table_data, colWidths=[2.2*inch, 3.8*inch])
         table.setStyle(TableStyle([
             ('BACKGROUND', (0,0), (-1,0), colors.HexColor('#0050BC')),
             ('TEXTCOLOR', (0,0), (-1,0), colors.white),
             ('ALIGN', (0,0), (-1,-1), 'LEFT'),
             ('VALIGN', (0,0), (-1,-1), 'MIDDLE'),
             ('FONTNAME', (0,0), (-1,0), 'Helvetica-Bold'),
-            ('FONTSIZE', (0,0), (-1,-1), 9),
-            ('BOTTOMPADDING', (0,0), (-1,0), 8),
-            ('TOPPADDING', (0,0), (-1,0), 8),
+            ('FONTSIZE', (0,0), (-1,-1), 8),
+            ('BOTTOMPADDING', (0,0), (-1,0), 6),
+            ('TOPPADDING', (0,0), (-1,0), 6),
             ('BACKGROUND', (0,1), (-1,-1), colors.HexColor('#F5F6FA')),
-            ('GRID', (0,0), (-1,-1), 0.5, colors.HexColor('#DDE4EB')),
+            ('GRID', (0,0), (-1,-1), 0.4, colors.HexColor('#DDE4EB')),
         ]))
         story.append(table)
-        story.append(Spacer(1, 0.4*inch))
+        story.append(Spacer(1, 0.3*inch))
         story.append(Paragraph("Prepared by: EvalBot - AI-Powered HR Analysis", body_text))
         story.append(PageBreak())
 
@@ -570,30 +570,30 @@ def create_pdf_report(analysis_data: Dict, output_path: str, gemini_report_text:
                 ['Confidence', voice_analysis.get('interpretation', {}).get('confidence_level', 'N/A'), f"Score: {voice_analysis.get('composite_scores', {}).get('confidence', 0):.3f}"],
                 ['Fluency', voice_analysis.get('interpretation', {}).get('fluency_level', 'N/A'), 'Drives engagement'],
             ]
-            table = Table(table_data, colWidths=[1.6*inch, 1.2*inch, 3.2*inch])
+            table = Table(table_data, colWidths=[1.5*inch, 1.3*inch, 3.2*inch])
             table.setStyle(TableStyle([
                 ('BACKGROUND', (0,0), (-1,0), colors.HexColor('#0050BC')),
                 ('TEXTCOLOR', (0,0), (-1,0), colors.white),
                 ('ALIGN', (0,0), (-1,-1), 'LEFT'),
                 ('VALIGN', (0,0), (-1,-1), 'MIDDLE'),
                 ('FONTNAME', (0,0), (-1,0), 'Helvetica-Bold'),
-                ('FONTSIZE', (0,0), (-1,-1), 9),
-                ('BOTTOMPADDING', (0,0), (-1,0), 8),
-                ('TOPPADDING', (0,0), (-1,0), 8),
+                ('FONTSIZE', (0,0), (-1,-1), 8),
+                ('BOTTOMPADDING', (0,0), (-1,0), 6),
+                ('TOPPADDING', (0,0), (-1,0), 6),
                 ('BACKGROUND', (0,1), (-1,-1), colors.HexColor('#F5F6FA')),
-                ('GRID', (0,0), (-1,-1), 0.5, colors.HexColor('#DDE4EB')),
+                ('GRID', (0,0), (-1,-1), 0.4, colors.HexColor('#DDE4EB')),
             ]))
             story.append(table)
-            story.append(Spacer(1, 0.2*inch))
+            story.append(Spacer(1, 0.15*inch))
             chart_buffer = io.BytesIO()
             generate_anxiety_confidence_chart(voice_analysis.get('composite_scores', {}), chart_buffer)
             chart_buffer.seek(0)
-            img = Image(chart_buffer, width=4.5*inch, height=3*inch)
+            img = Image(chart_buffer, width=4.2*inch, height=2.8*inch)
             img.hAlign = 'CENTER'
             story.append(img)
         else:
-            story.append(Paragraph("Vocal analysis unavailable.", body_text))
-        story.append(Spacer(1, 0.2*inch))
+            story.append(Paragraph("Vocal analysis unavailable due to processing limitations.", body_text))
+        story.append(Spacer(1, 0.15*inch))
 
         # Parse Gemini Report
         sections = {
@@ -609,7 +609,6 @@ def create_pdf_report(analysis_data: Dict, output_path: str, gemini_report_text:
         for line in lines:
             line = line.strip()
             if not line: continue
-            # Simplified regex to avoid parenthesis issues
             if line.startswith('**') and line.endswith('**'):
                 section_title = line.strip('**').strip()
                 if section_title.startswith(('1.', '2.', '3.', '4.', '5.')):
@@ -632,7 +631,7 @@ def create_pdf_report(analysis_data: Dict, output_path: str, gemini_report_text:
             elif line.startswith(('-', '*', '•')) and current_section:
                 clean_line = line.lstrip('-*• ').strip()
                 if not clean_line: continue
-                clean_line = re.sub(r'[()]', '', clean_line)  # Remove parentheses
+                clean_line = re.sub(r'[()]', '', clean_line)
                 if current_section == 'Competency':
                     if any(k in clean_line.lower() for k in ['leader', 'problem', 'commun', 'adapt', 'strength']):
                         current_subsection = 'Strengths'
@@ -657,10 +656,10 @@ def create_pdf_report(analysis_data: Dict, output_path: str, gemini_report_text:
                 story.append(Paragraph(line, bullet_style))
         else:
             story.append(Paragraph("No summary provided.", body_text))
-        story.append(Spacer(1, 0.2*inch))
+        story.append(Spacer(1, 0.15*inch))
 
         # Competency and Content
-        story.append(Paragraph("3. Competency & Evaluation", h2))
+        story.append(Paragraph("3. Competency & Content", h2))
         story.append(Paragraph("Strengths", h3))
         if sections['Competency']['Strengths']:
             for line in sections['Competency']['Strengths']:
@@ -674,7 +673,7 @@ def create_pdf_report(analysis_data: Dict, output_path: str, gemini_report_text:
                 story.append(Paragraph(line, bullet_style))
         else:
             story.append(Paragraph("No growth areas identified; maintain current strengths.", body_text))
-        story.append(Spacer(1, 0.2*inch))
+        story.append(Spacer(1, 0.15*inch))
 
         # Role Fit
         story.append(Paragraph("4. Role Fit & Potential", h2))
@@ -683,7 +682,7 @@ def create_pdf_report(analysis_data: Dict, output_path: str, gemini_report_text:
                 story.append(Paragraph(line, bullet_style))
         else:
             story.append(Paragraph("No fit analysis provided.", body_text))
-        story.append(Spacer(1, 0.2*inch))
+        story.append(Spacer(1, 0.15*inch))
 
         # Recommendations
         story.append(Paragraph("5. Recommendations", h2))
@@ -694,13 +693,14 @@ def create_pdf_report(analysis_data: Dict, output_path: str, gemini_report_text:
         else:
             story.append(Paragraph("No development priorities specified.", body_text))
         story.append(Spacer(1, 0.1*inch))
-        story.append(Paragraph("Next Steps", h3))
+        story.append(Paragraph("Next Steps for Hiring Managers", h3))
         if sections['Recommendations']['Next Steps']:
             for line in sections['Recommendations']['Next Steps']:
                 story.append(Paragraph(line, bullet_style))
         else:
             story.append(Paragraph("No next steps provided.", body_text))
-        story.append(Spacer(1, 0.2*inch))
+        story.append(Spacer(1, 0.15*inch))
+        story.append(Paragraph("This report provides actionable insights to support hiring and candidate development.", body_text))
 
         doc.build(story, onFirstPage=header_footer, onLaterPages=header_footer)
         logger.info(f"PDF report successfully generated at {output_path}")
@@ -716,17 +716,20 @@ def convert_to_serializable(obj):
     if isinstance(obj, np.ndarray): return obj.tolist()
     return obj
 
-def process_interview(audio_path_or_url: str) -> Dict:
+def process_interview(audio_url: str) -> Dict:
+    """Process a single audio URL and generate analysis report."""
     local_audio_path = None
     wav_file = None
     is_downloaded = False
     try:
-        logger.info(f"Starting processing for {audio_path_or_url}")
-        if audio_path_or_url.startswith(('http://', 'https://')):
-            local_audio_path = download_audio_from_url(audio_path_or_url)
+        if not isinstance(audio_url, str):
+            raise ValueError("Input must be a single URL string")
+        logger.info(f"Starting processing for {audio_url}")
+        if audio_url.startswith(('http://', 'https://')):
+            local_audio_path = download_audio_from_url(audio_url)
             is_downloaded = True
         else:
-            local_audio_path = audio_path_or_url
+            local_audio_path = audio_url
             if not os.path.exists(local_audio_path):
                 raise FileNotFoundError(f"Local audio file not found: {local_audio_path}")
         wav_file = convert_to_wav(local_audio_path)
@@ -762,16 +765,16 @@ def process_interview(audio_path_or_url: str) -> Dict:
             serializable_data = convert_to_serializable(analysis_data)
             json.dump(serializable_data, f, indent=2)
         if not pdf_success:
-            logger.warning(f"PDF report failed to generate for {audio_path_or_url}")
+            logger.warning(f"PDF report failed to generate for {audio_url}")
             return {
                 'pdf_path': None,
                 'json_path': json_path,
                 'error': 'PDF generation failed'
             }
-        logger.info(f"Processing completed for {audio_path_or_url}")
+        logger.info(f"Processing completed for {audio_url}")
         return {'pdf_path': pdf_path, 'json_path': json_path}
     except Exception as e:
-        logger.error(f"Processing failed for {audio_path_or_url}: {str(e)}", exc_info=True)
+        logger.error(f"Processing failed for {audio_url}: {str(e)}", exc_info=True)
         base_name = str(uuid.uuid4())
         json_path = os.path.join(OUTPUT_DIR, f"{base_name}_analysis.json")
         with open(json_path, 'w') as f: