Upload casl_analysis.py

casl_analysis.py

@@ -1,4 +1,4 @@
-import gradio as gr
+eimport gradio as gr
 import boto3
 import json
 import pandas as pd
@@ -11,9 +11,31 @@ import pickle
 import csv
 from PIL import Image
 import io
-import PyPDF2
 import uuid
 from datetime import datetime
+import tempfile
+import time
+
+# Try to import ReportLab (needed for PDF generation)
+try:
+    from reportlab.lib.pagesizes import letter
+    from reportlab.lib import colors
+    from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, Table, TableStyle
+    from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
+    REPORTLAB_AVAILABLE = True
+except ImportError:
+    logger = logging.getLogger(__name__)
+    logger.warning("ReportLab library not available - PDF export will be disabled")
+    REPORTLAB_AVAILABLE = False
+
+# Try to import PyPDF2 (needed for PDF reading)
+try:
+    import PyPDF2
+    PYPDF2_AVAILABLE = True
+except ImportError:
+    logger = logging.getLogger(__name__)
+    logger.warning("PyPDF2 library not available - PDF reading will be disabled")
+    PYPDF2_AVAILABLE = False
 
 # Configure logging
 logging.basicConfig(level=logging.INFO)
@@ -25,10 +47,14 @@ AWS_ACCESS_KEY = os.getenv("AWS_ACCESS_KEY", "")
 AWS_SECRET_KEY = os.getenv("AWS_SECRET_KEY", "")
 AWS_REGION = os.getenv("AWS_REGION", "us-east-1")
 
-# Initialize Bedrock client if credentials are available
+# Initialize AWS clients if credentials are available
 bedrock_client = None
+transcribe_client = None
+s3_client = None
+
 if AWS_ACCESS_KEY and AWS_SECRET_KEY:
     try:
+        # Initialize Bedrock client for AI analysis
         bedrock_client = boto3.client(
             'bedrock-runtime',
             aws_access_key_id=AWS_ACCESS_KEY,
@@ -36,8 +62,30 @@ if AWS_ACCESS_KEY and AWS_SECRET_KEY:
             region_name=AWS_REGION
         )
         logger.info("Bedrock client initialized successfully")
+        
+        # Initialize Transcribe client for speech-to-text
+        transcribe_client = boto3.client(
+            'transcribe',
+            aws_access_key_id=AWS_ACCESS_KEY,
+            aws_secret_access_key=AWS_SECRET_KEY,
+            region_name=AWS_REGION
+        )
+        logger.info("Transcribe client initialized successfully")
+        
+        # Initialize S3 client for storing audio files
+        s3_client = boto3.client(
+            's3',
+            aws_access_key_id=AWS_ACCESS_KEY,
+            aws_secret_access_key=AWS_SECRET_KEY,
+            region_name=AWS_REGION
+        )
+        logger.info("S3 client initialized successfully")
     except Exception as e:
-        logger.error(f"Failed to initialize Bedrock client: {str(e)}")
+        logger.error(f"Failed to initialize AWS clients: {str(e)}")
+
+# S3 bucket for storing audio files
+S3_BUCKET = os.environ.get("S3_BUCKET", "casl-audio-files") 
+S3_PREFIX = "transcribe-audio/"
 
 # Sample transcript for the demo
 SAMPLE_TRANSCRIPT = """*PAR: today I would &-um like to talk about &-um a fun trip I took last &-um summer with my family.
@@ -60,23 +108,38 @@ SAMPLE_TRANSCRIPT = """*PAR: today I would &-um like to talk about &-um a fun tr
 # ===============================
 
 # Create data directories if they don't exist
-DATA_DIR = "patient_data"
+DATA_DIR = os.environ.get("DATA_DIR", "patient_data")
 RECORDS_FILE = os.path.join(DATA_DIR, "patient_records.csv")
 ANALYSES_DIR = os.path.join(DATA_DIR, "analyses")
+DOWNLOADS_DIR = os.path.join(DATA_DIR, "downloads")
+AUDIO_DIR = os.path.join(DATA_DIR, "audio")
 
 def ensure_data_dirs():
     """Ensure data directories exist"""
-    os.makedirs(DATA_DIR, exist_ok=True)
-    os.makedirs(ANALYSES_DIR, exist_ok=True)
-    
-    # Create records file if it doesn't exist
-    if not os.path.exists(RECORDS_FILE):
-        with open(RECORDS_FILE, 'w', newline='') as f:
-            writer = csv.writer(f)
-            writer.writerow([
-                "ID", "Name", "Record ID", "Age", "Gender", 
-                "Assessment Date", "Clinician", "Analysis Date", "File Path"
-            ])
+    global DOWNLOADS_DIR, AUDIO_DIR
+    try:
+        os.makedirs(DATA_DIR, exist_ok=True)
+        os.makedirs(ANALYSES_DIR, exist_ok=True)
+        os.makedirs(DOWNLOADS_DIR, exist_ok=True)
+        os.makedirs(AUDIO_DIR, exist_ok=True)
+        logger.info(f"Data directories created: {DATA_DIR}, {ANALYSES_DIR}, {DOWNLOADS_DIR}, {AUDIO_DIR}")
+        
+        # Create records file if it doesn't exist
+        if not os.path.exists(RECORDS_FILE):
+            with open(RECORDS_FILE, 'w', newline='') as f:
+                writer = csv.writer(f)
+                writer.writerow([
+                    "ID", "Name", "Record ID", "Age", "Gender", 
+                    "Assessment Date", "Clinician", "Analysis Date", "File Path"
+                ])
+    except Exception as e:
+        logger.warning(f"Could not create data directories: {str(e)}")
+        # Fallback to tmp directory on HF Spaces
+        DOWNLOADS_DIR = os.path.join(tempfile.gettempdir(), "casl_downloads")
+        AUDIO_DIR = os.path.join(tempfile.gettempdir(), "casl_audio")
+        os.makedirs(DOWNLOADS_DIR, exist_ok=True)
+        os.makedirs(AUDIO_DIR, exist_ok=True)
+        logger.info(f"Using fallback directories: {DOWNLOADS_DIR}, {AUDIO_DIR}")
 
 # Initialize data directories
 ensure_data_dirs()
@@ -278,6 +341,9 @@ def delete_patient_record(record_id):
 
 def read_pdf(file_path):
     """Read text from a PDF file"""
+    if not PYPDF2_AVAILABLE:
+        return "Error: PDF reading is not available - PyPDF2 library is not installed"
+    
     try:
         with open(file_path, 'rb') as file:
             pdf_reader = PyPDF2.PdfReader(file)
@@ -318,7 +384,10 @@ def process_upload(file):
     
     file_path = file.name
     if file_path.endswith('.pdf'):
-        return read_pdf(file_path)
+        if PYPDF2_AVAILABLE:
+            return read_pdf(file_path)
+        else:
+            return "Error: PDF reading is disabled - PyPDF2 library is not installed"
     elif file_path.endswith('.cha'):
         return read_cha_file(file_path)
     else:
@@ -424,10 +493,15 @@ def generate_demo_response(prompt):
     
     return response
 
-def generate_demo_transcription(audio_path):
+def generate_demo_transcription():
     """Generate a simulated transcription response"""
-    # In a real app, this would process an audio file
-    return "*PAR: today I want to tell you about my favorite toy.\n*PAR: it's a &-um teddy bear that I got for my birthday.\n*PAR: he has &-um brown fur and a red bow.\n*PAR: I like to sleep with him every night.\n*PAR: sometimes I take him to school in my backpack."
+    return """*PAR: today I want to tell you about my favorite toy.
+*PAR: it's a &-um teddy bear that I got for my birthday.
+*PAR: he has &-um brown fur and a red bow.
+*PAR: I like to sleep with him every night.
+*PAR: sometimes I take him to school in my backpack.
+*INV: what's your teddy bear's name?
+*PAR: his name is &-um Brownie because he's brown."""
 
 def generate_demo_qa_response(question):
     """Generate a simulated Q&A response"""
@@ -877,109 +951,302 @@ def analyze_transcript(transcript, age, gender):
     
     # Instructions for the LLM analysis
     instructions = """
-    You are a speech pathologist analyzing this transcription sample. Provide a detailed analysis focused on specific quotes from the transcript.
-
-    The factors of speech that you need to count are:
-
-    1. Difficulty producing fluent, grammatical speech - speech that is slow, halting, with pauses while searching for words
-    2. Word retrieval issues - trouble thinking of specific words, use of filler words like um, circumlocution, semantically similar word substitutions
-    3. Grammatical errors - missing/incorrect function words, problems with verb tenses, conjugation, agreement, simplified sentences
-    4. Repetitions and revisions - repeating or restating words, phrases or sentences due to trouble finding the right words
-    5. Neologisms - creating nonexistent "new" words
-    6. Perseveration - unintentionally repeating words or phrases over and over
-    7. Comprehension issues - trouble understanding complex sentences, fast speech, relying more on context and cues
-
-    For each factor, provide:
-    - Number of occurrences
-    - Severity percentile (estimate based on your clinical judgment)
-    - At least 2-3 specific quotes from the transcript as examples
-
-    Then evaluate using the CASL-2 Speech and Language Analysis Framework across these domains:
-
-    1. Lexical/Semantic Skills:
-       - Assess vocabulary diversity, word-finding abilities, semantic precision
-       - Provide Standard Score (mean=100, SD=15), percentile rank, and performance level
-       - Include SPECIFIC QUOTES as evidence
+    Advanced Linguistic Analysis Protocol for Adolescent Language Samples (Ages 14-18)
+You are a highly specialized assistant supporting speech-language pathologists in conducting comprehensive linguistic analyses of adolescent language samples. Your analysis must adhere to evidence-based practice standards for secondary-level language assessment while producing results that inform both clinical decision-making and family understanding.
 
-    2. Syntactic Skills:
-       - Evaluate grammatical accuracy, sentence complexity, morphological skills
-       - Provide Standard Score, percentile rank, and performance level
-       - Include SPECIFIC QUOTES as evidence
+Initial Sample Assessment
+1. Document sample metadata:
+    * Total number of utterances
+    * Sample collection context (conversational, narrative, expository, persuasive, procedural)
+    * Sample elicitation method (if indicated)
+    * Total duration/length of interaction
+    * Any transcription conventions used (e.g., SALT, CHAT)
+2. Determine analysis approach:
+    * For samples with ≤50 utterances: Perform complete analysis on all utterances
+    * For samples >50 utterances: Perform complete analysis, then select 50 representative utterances that capture key patterns across all linguistic domains while maintaining the natural distribution of features
+Utterance-Level Microanalysis
+For each utterance, provide detailed linguistic breakdown including:
+1. Structural Components
+    * Utterance number and full text (verbatim)
+    * Word count (excluding fillers, false starts, and repetitions)
+    * C-unit segmentation (when applicable)
+    * MLU in words and morphemes
+    * Syntactic classification:
+        * Simple, compound, complex, or compound-complex
+        * Complete or fragmentary
+        * Declarative, interrogative, imperative, or exclamatory
+    * Clausal density (number of clauses/utterance)
+2. Syntactic Analysis
+    * Constituent structure identification:
+        * Subject and predicate components
+        * Noun phrases (including pre- and post-modification)
+        * Verb phrases (including auxiliaries, complements)
+        * Adverbial phrases and clauses (including position and function)
+        * Prepositional phrases (including syntactic role)
+        * Subordinate clauses (including type and function)
+    * Embedding depth and recursion patterns
+    * Syntactic movement and transformations
+    * Non-canonical structures (passives, clefts, etc.)
+3. Morphological Analysis
+    * Bound morpheme usage (inflectional and derivational)
+    * Tense marking consistency
+    * Agreement patterns (subject-verb, pronoun-antecedent)
+    * Morphological errors with classification
+4. Error Analysis (for each identified error)
+    * Precise error location:
+        * Utterance number and full quotation
+        * Position within utterance (initial, medial, final)
+        * Syntactic position (e.g., main clause, subordinate clause, noun phrase)
+        * Proximity to other linguistic features (e.g., complex vocabulary, disfluencies)
+    * Error type classification:
+        * Morphosyntactic (agreement, tense, etc.)
+        * Lexical-semantic (word selection, collocational)
+        * Phonological (if transcribed)
+        * Pragmatic (if contextually inappropriate)
+    * Error pattern (developmental vs. atypical)
+    * Clinical significance of location:
+        * Relationship to sentence complexity (errors increasing with complexity)
+        * Patterns related to linguistic context (e.g., errors occurring after disfluencies)
+        * Consistency across similar syntactic environments
+        * Relationship to cognitive load (e.g., errors increasing in dense information units)
+    * Error frequency and distribution across contexts
+    * Self-correction attempts and success rate
+5. Fluency Markers
+    * Mazes (false starts, repetitions, reformulations)
+    * Filled pauses (um, uh, like, etc.)
+    * Silent pauses (duration if indicated)
+    * Disruption patterns and positions
+    * Impact on communicative effectiveness
+QUANTITATIVE ANALYSIS & METRICS
+Calculate the following evidence-based metrics with interpretations relevant to adolescent language development:
+Productivity Measures
+1. Total Output Measures
+    * Total number of words (TNW)
+    * Total number of utterances (TNU)
+    * Total number of different words (TDW)
+    * Total communication units (T-units/C-units)
+2. Length Measures
+    * Mean length of utterance in words (MLU-w)
+    * Mean length of utterance in morphemes (MLU-m)
+    * Mean length of C-unit (MLCU)
+    * Words per minute (if timing available)
+Complexity Measures
+1. Syntactic Complexity
+    * Clausal density (clauses per C-unit)
+    * Subordination index (SI)
+    * Coordination index
+    * Embedding depth (max levels of embedding)
+    * T-unit complexity ratio
+    * Percentage of complex sentences
+2. Phrase-Level Complexity
+    * Mean noun phrase length
+    * Mean verb phrase complexity
+    * Prepositional phrase frequency
+    * Adverbial complexity
+Accuracy Measures
+1. Error Analysis Summary
+    * Percentage of grammatically correct utterances
+    * Errors per C-unit
+    * Error pattern distribution (morphological, syntactic, lexical)
+    * Most frequent error types with specific examples
+    * Error location patterns:
+        * Distribution across utterance positions (initial, medial, final)
+        * Distribution across syntactic structures (simple vs. complex)
+        * Correlation with utterance length and complexity
+        * Patterns related to information density or processing demands
+    * Clinical significance of error locations:
+        * Interpretation of position-specific error patterns
+        * Analysis of syntactic contexts where errors predominate
+        * Relationship between error location and communicative impact
+Lexical Diversity & Sophistication
+1. Vocabulary Metrics
+    * Type-token ratio (TTR)
+    * Moving-average type-token ratio (MATTR)
+    * Number of different words (NDW)
+    * Vocabulary diversity (D)
+    * Lexical density (content words/total words)
+2. Lexical Sophistication
+    * Low-frequency word usage
+    * Academic vocabulary presence
+    * Abstract word usage
+    * Word specificity analysis
+Fluency & Formulation Measures
+1. Disruption Analysis
+    * Percentage of mazes
+    * Total maze words/total words
+    * Revisions per utterance
+    * Hesitation frequency
+    * Incomplete utterance percentage
+    * Word-finding difficulties (frequency and patterns)
+CASL-2 DOMAIN ALIGNMENT
+Analyze the sample according to the Comprehensive Assessment of Spoken Language (CASL-2) framework, providing detailed evidence for each domain:
+1. Lexical/Semantic Domain
+* Vocabulary Range Assessment
+    * Basic vs. precise vocabulary usage
+    * Abstract vs. concrete terminology
+    * Academic language presence
+    * Subject-specific terminology
+    * Register-appropriate lexicon
+* Word Relationships
+    * Synonym/antonym usage
+    * Categorical relationships
+    * Part-whole relationships
+    * Semantic networks
+* Word Retrieval Patterns
+    * Word-finding hesitations
+    * Circumlocutions
+    * Semantic substitutions
+    * Retrieval strategies
+* Evidence Summary
+    * Provide specific examples from transcript
+    * Compare to age-appropriate expectations
+    * Estimate standard score range (using clinical judgment)
+    * Indicate percentile rank range
+    * Assign performance level category
+2. Syntactic Domain
+* Sentence Structure Analysis
+    * Distribution of sentence types
+    * Complex syntax usage patterns
+    * Syntactic versatility
+    * Age-appropriate structures
+* Morphosyntactic Elements
+    * Regular and irregular morphology
+    * Verb tense system mastery
+    * Complex verb forms (perfect, progressive)
+    * Advanced agreement patterns
+* Syntactic Maturity Indicators
+    * Clause combining strategies
+    * Embedding types and frequency
+    * Noun phrase elaboration
+    * Adverbial complexity
+* Evidence Summary
+    * Provide specific examples from transcript
+    * Compare to age-appropriate expectations
+    * Estimate standard score range
+    * Indicate percentile rank range
+    * Assign performance level category
+3. Supralinguistic Domain
+* Figurative Language
+    * Idiomatic expressions
+    * Metaphors and analogies
+    * Humor or wordplay
+    * Understanding of non-literal content
+* Higher-Order Reasoning
+    * Inferential language
+    * Ambiguity recognition
+    * Abstract concept expression
+    * Perspective-taking indicators
+* Metalinguistic Awareness
+    * Self-monitoring
+    * Linguistic reflection
+    * Awareness of language rules
+    * Metacognitive comments
+* Evidence Summary
+    * Provide specific examples from transcript
+    * Note limitations of assessment from sample
+    * Estimate standard score range (if sufficient evidence)
+    * Indicate percentile rank range (if applicable)
+    * Assign performance level category (or note insufficient evidence)
+4. Pragmatic Domain
+* Discourse Management
+    * Topic initiation, maintenance, and change
+    * Turn-taking patterns
+    * Response contingency
+    * Conversational repair strategies
+* Social Communication
+    * Perspective-taking
+    * Register variation
+    * Politeness conventions
+    * Social inferencing
+* Narrative/Expository Skills (if applicable)
+    * Coherence and cohesion
+    * Organizational structure
+    * Use of cohesive devices
+    * Information density
+* Evidence Summary
+    * Provide specific examples from transcript
+    * Note contextual limitations
+    * Estimate standard score range (if sufficient evidence)
+    * Indicate percentile rank range (if applicable)
+    * Assign performance level category (or note insufficient evidence)
+DEVELOPMENTAL PROFILE ANALYSIS
+Compare observed language features to established adolescent language development patterns:
+1. Age-Based Comparison
+    * Alignment with typical syntactic development (14-18)
+    * Lexical development expectations
+    * Discourse maturity indicators
+    * Academic language benchmarks
+2. Strength-Challenge Pattern Analysis
+    * Identify domains of relative strength with evidence
+    * Identify domains requiring support with evidence
+    * Note any asynchronous development patterns
+    * Document compensatory strategies observed
+3. Developmental Trajectory Indicators
+    * Features suggesting typical development
+    * Features suggesting delayed development
+    * Features suggesting disordered development
+    * Features suggesting language difference vs. disorder
+COMPREHENSIVE REPORTING FORMAT
+1. Professional Clinical Summary (SLP-Oriented)
+* Sample characteristics and analysis methodology
+* Key quantitative findings table with age-based interpretation
+* CASL-2 domain profiles with evidence-based rationales
+* Error pattern analysis with clinical implications
+* Identified strengths and challenges
+* Differential considerations
+* Recommendations for further assessment
+* Potential treatment targets based on evidence
+2. Family-Friendly Summary Report
+* Introduction
+    * Purpose of language sample analysis
+    * Brief explanation of what was analyzed
+    * How this information helps understand communication
+* Your Adolescent's Language Profile
+    * Overall communication strengths (with clear examples)
+    * Areas for continued growth (with supportive examples)
+    * How these patterns may impact academic and social communication
+* Understanding the Assessment
+    * Simple explanations of key findings
+    * Comparison to typical adolescent language patterns
+    * Visual representation of language profile
+    * Accessible examples from the transcript
+* Supporting Language Development
+    * Practical strategies aligned with findings
+    * Communication opportunities that leverage strengths
+    * Questions to discuss with the SLP
+    * Resources for family understanding
+* Next Steps
+    * Connections to academic and social communication
+    * Relevance to current educational goals
+    * Partnership opportunities between home and therapy
+3. Educational Implications (if requested)
+* Connections to academic standards
+* Impact on classroom participation
+* Alignment with IEP goals (if applicable)
+* Recommendations for classroom support
+IMPLEMENTATION GUIDELINES
+1. Analysis Integrity
+    * Analyze only what is directly observable in the transcript
+    * Clearly differentiate observations from interpretations
+    * Note when certain domains cannot be adequately assessed
+    * Document analysis limitations based on sample constraints
+2. Clinical Reasoning
+    * Apply evidence-based standards for adolescent language
+    * Consider developmental appropriateness for ages 14-18
+    * Document patterns rather than isolated instances
+    * Provide context for interpretations
+3. Reporting Ethics
+    * Use person-first, strength-based language
+    * Avoid definitive diagnostic statements
+    * Focus on functional communication impact
+    * Maintain appropriate scope of analysis
+4. Flexibility Adaptations
+    * For different discourse types (narrative, expository, conversational)
+    * For different cultural and linguistic backgrounds
+    * For various academic and social contexts
+    * For potential co-occurring conditions
+This protocol produces a comprehensive linguistic analysis tailored to adolescents (14-18) that provides both clinically relevant information and family-accessible insights while maintaining the flexibility to adapt to various sample types and contexts.
 
-    3. Supralinguistic Skills:
-       - Assess figurative language use, inferencing, and abstract reasoning
-       - Provide Standard Score, percentile rank, and performance level
-       - Include SPECIFIC QUOTES as evidence
-
-    YOUR RESPONSE MUST USE THESE EXACT SECTION MARKERS FOR PARSING:
-
-    <SPEECH_FACTORS_START>
-    Difficulty producing fluent, grammatical speech: (occurrences), (percentile)
-    Examples:
-    - "(direct quote from transcript)"
-    - "(direct quote from transcript)"
-    
-    Word retrieval issues: (occurrences), (percentile)
-    Examples:
-    - "(direct quote from transcript)"
-    - "(direct quote from transcript)"
-    
-    (And so on for each factor)
-    <SPEECH_FACTORS_END>
-    
-    <CASL_SKILLS_START>
-    Lexical/Semantic Skills: Standard Score (X), Percentile Rank (X%), Performance Level
-    Examples:
-    - "(direct quote showing strength or weakness)"
-    - "(direct quote showing strength or weakness)"
-    
-    Syntactic Skills: Standard Score (X), Percentile Rank (X%), Performance Level
-    Examples:
-    - "(direct quote showing strength or weakness)"
-    - "(direct quote showing strength or weakness)"
-    
-    Supralinguistic Skills: Standard Score (X), Percentile Rank (X%), Performance Level
-    Examples:
-    - "(direct quote showing strength or weakness)"
-    - "(direct quote showing strength or weakness)"
-    <CASL_SKILLS_END>
-    
-    <TREATMENT_RECOMMENDATIONS_START>
-    - (treatment recommendation)
-    - (treatment recommendation)
-    - (treatment recommendation)
-    <TREATMENT_RECOMMENDATIONS_END>
-    
-    <EXPLANATION_START>
-    (brief diagnostic rationale based on findings)
-    <EXPLANATION_END>
-    
-    <ADDITIONAL_ANALYSIS_START>
-    (specific insights that would be helpful for treatment planning)
-    <ADDITIONAL_ANALYSIS_END>
-    
-    <DIAGNOSTIC_IMPRESSIONS_START>
-    (summarize findings across domains using specific examples and clear explanations)
-    <DIAGNOSTIC_IMPRESSIONS_END>
-    
-    <ERROR_EXAMPLES_START>
-    (Copy all the specific quote examples here again, organized by error type or skill domain)
-    <ERROR_EXAMPLES_END>
-    
-    MOST IMPORTANT: 
-    1. Use EXACTLY the section markers provided (like <SPEECH_FACTORS_START>) to make parsing reliable
-    2. For EVERY factor and domain you analyze, you MUST provide direct quotes from the transcript as evidence
-    3. Be very specific and cite the exact text
-    4. Do not omit any of the required sections
-    """
-    
-    # Prepare prompt for Claude with the user's role context
-    role_context = """
-    You are a speech pathologist, a healthcare professional who specializes in evaluating, diagnosing, and treating communication disorders, including speech, language, cognitive-communication, voice, swallowing, and fluency disorders. Your role is to help patients improve their speech and communication skills through various therapeutic techniques and exercises.
-
-    You are working with a student with speech impediments.
-
-    The most important thing is that you stay kind to the child. Be constructive and helpful rather than critical.
     """
     
     prompt = f"""
@@ -1013,19 +1280,190 @@ def analyze_transcript(transcript, age, gender):
     return results, plot_image, radar_image, response
 
 
-def transcribe_audio(audio_path, patient_age):
-    """Transcribe an audio recording using CHAT format"""
-    # In a real implementation, this would use a speech-to-text service
-    # For demo purposes, we'll return a simulated transcription
-    
-    if bedrock_client:
-        # In a real implementation, you would process the audio file and send it to a transcription service
-        # Here we just simulate the result
-        transcription = generate_demo_transcription(audio_path)
-    else:
-        transcription = generate_demo_transcription(audio_path)
-    
-    return transcription
+def transcribe_audio(audio_path, patient_age=8):
+    """Transcribe an audio recording using Amazon Transcribe and format in CHAT format"""
+    if not os.path.exists(audio_path):
+        logger.error(f"Audio file not found: {audio_path}")
+        return "Error: Audio file not found."
+        
+    if not transcribe_client or not s3_client:
+        logger.warning("AWS clients not initialized, using demo transcription")
+        return generate_demo_transcription()
+        
+    try:
+        # Get file info
+        file_name = os.path.basename(audio_path)
+        file_size = os.path.getsize(audio_path)
+        _, file_extension = os.path.splitext(file_name)
+        
+        # Check file format
+        supported_formats = ['.mp3', '.mp4', '.wav', '.flac', '.ogg', '.amr', '.webm']
+        if file_extension.lower() not in supported_formats:
+            logger.error(f"Unsupported audio format: {file_extension}")
+            return f"Error: Unsupported audio format. Please use one of: {', '.join(supported_formats)}"
+            
+        # Generate a unique job name
+        timestamp = datetime.now().strftime('%Y%m%d%H%M%S')
+        job_name = f"casl-transcription-{timestamp}"
+        s3_key = f"{S3_PREFIX}{job_name}{file_extension}"
+        
+        # Upload to S3
+        logger.info(f"Uploading {file_name} to S3 bucket {S3_BUCKET}")
+        try:
+            with open(audio_path, 'rb') as audio_file:
+                s3_client.upload_fileobj(audio_file, S3_BUCKET, s3_key)
+        except Exception as e:
+            logger.error(f"Failed to upload to S3: {str(e)}")
+            
+            # If upload fails, try to create the bucket
+            try:
+                s3_client.create_bucket(Bucket=S3_BUCKET)
+                logger.info(f"Created S3 bucket: {S3_BUCKET}")
+                
+                # Try upload again
+                with open(audio_path, 'rb') as audio_file:
+                    s3_client.upload_fileobj(audio_file, S3_BUCKET, s3_key)
+            except Exception as bucket_error:
+                logger.error(f"Failed to create bucket and upload: {str(bucket_error)}")
+                return "Error: Failed to upload audio file. Please check your AWS permissions."
+        
+        # Start transcription job
+        logger.info(f"Starting transcription job: {job_name}")
+        media_format = file_extension.lower()[1:]  # Remove the dot
+        if media_format == 'webm':
+            media_format = 'webm'  # Amazon Transcribe expects this
+            
+        # Determine language settings based on patient age
+        if patient_age < 10:
+            # For younger children, enabling child language model is helpful
+            language_options = {
+                'LanguageCode': 'en-US',
+                'Settings': {
+                    'ShowSpeakerLabels': True,
+                    'MaxSpeakerLabels': 2  # Typically patient + clinician
+                }
+            }
+        else:
+            language_options = {
+                'LanguageCode': 'en-US',
+                'Settings': {
+                    'ShowSpeakerLabels': True,
+                    'MaxSpeakerLabels': 2  # Typically patient + clinician
+                }
+            }
+            
+        transcribe_client.start_transcription_job(
+            TranscriptionJobName=job_name,
+            Media={
+                'MediaFileUri': f"s3://{S3_BUCKET}/{s3_key}"
+            },
+            MediaFormat=media_format,
+            **language_options
+        )
+        
+        # Wait for the job to complete (with timeout)
+        logger.info("Waiting for transcription to complete...")
+        max_tries = 30  # 5 minutes max wait
+        tries = 0
+        
+        while tries < max_tries:
+            try:
+                job = transcribe_client.get_transcription_job(TranscriptionJobName=job_name)
+                status = job['TranscriptionJob']['TranscriptionJobStatus']
+                
+                if status == 'COMPLETED':
+                    # Get the transcript
+                    transcript_uri = job['TranscriptionJob']['Transcript']['TranscriptFileUri']
+                    
+                    # Download the transcript
+                    import urllib.request
+                    import json
+                    
+                    with urllib.request.urlopen(transcript_uri) as response:
+                        transcript_json = json.loads(response.read().decode('utf-8'))
+                    
+                    # Convert to CHAT format
+                    chat_transcript = format_as_chat(transcript_json)
+                    return chat_transcript
+                    
+                elif status == 'FAILED':
+                    reason = job['TranscriptionJob'].get('FailureReason', 'Unknown failure')
+                    logger.error(f"Transcription job failed: {reason}")
+                    return f"Error: Transcription failed - {reason}"
+                    
+                # Still in progress, wait and try again
+                tries += 1
+                time.sleep(10)  # Check every 10 seconds
+                
+            except Exception as e:
+                logger.error(f"Error checking transcription job: {str(e)}")
+                return f"Error getting transcription: {str(e)}"
+        
+        # If we got here, we timed out
+        return "Error: Transcription timed out. The process is taking longer than expected."
+        
+    except Exception as e:
+        logger.exception("Error in audio transcription")
+        return f"Error transcribing audio: {str(e)}"
+
+def format_as_chat(transcript_json):
+    """Format the Amazon Transcribe JSON result as CHAT format"""
+    try:
+        # Get transcript items
+        items = transcript_json['results']['items']
+        
+        # Get speaker labels if available
+        speakers = {}
+        if 'speaker_labels' in transcript_json['results']:
+            speaker_segments = transcript_json['results']['speaker_labels']['segments']
+            
+            # Map each item to its speaker
+            for segment in speaker_segments:
+                for item in segment['items']:
+                    start_time = item['start_time']
+                    speakers[start_time] = segment['speaker_label']
+        
+        # Build transcript by combining words into utterances by speaker
+        current_speaker = None
+        current_utterance = []
+        utterances = []
+        
+        for item in items:
+            # Skip non-pronunciation items (like punctuation)
+            if item['type'] != 'pronunciation':
+                continue
+                
+            word = item['alternatives'][0]['content']
+            start_time = item.get('start_time')
+            
+            # Determine speaker if available
+            speaker = speakers.get(start_time, 'spk_0')
+            
+            # If speaker changed, start a new utterance
+            if speaker != current_speaker and current_utterance:
+                utterances.append((current_speaker, ' '.join(current_utterance)))
+                current_utterance = []
+                
+            current_speaker = speaker
+            current_utterance.append(word)
+        
+        # Add the last utterance
+        if current_utterance:
+            utterances.append((current_speaker, ' '.join(current_utterance)))
+        
+        # Format as CHAT
+        chat_lines = []
+        for speaker, text in utterances:
+            # Map speakers to CHAT format
+            # Assuming spk_0 is the patient (PAR) and spk_1 is the clinician (INV)
+            chat_speaker = "*PAR:" if speaker == "spk_0" else "*INV:"
+            chat_lines.append(f"{chat_speaker} {text}.")
+            
+        return '\n'.join(chat_lines)
+        
+    except Exception as e:
+        logger.exception("Error formatting transcript")
+        return "*PAR: (Error formatting transcript)"
 
 def answer_slp_question(question):
     """Answer a question about SLP practice or CASL assessment"""
@@ -1334,12 +1772,8 @@ def create_interface():
                     outputs=[patient_records_table, records_status]
                 )
                 
-                # Automatically load records when tab is selected
-                main_tabs.select(
-                    lambda tab_id: refresh_patient_records() if tab_id == 1 else (pd.DataFrame(), ""),
-                    inputs=[main_tabs],
-                    outputs=[patient_records_table, records_status]
-                )
+                # Note: The automatic tab selection event was removed because it's not supported in newer Gradio versions
+                # Instead, we'll rely on the refresh button that's already in place
                 
                 # Load record when a row is selected
                 def handle_record_selection(evt: gr.SelectData, records):
@@ -1737,15 +2171,11 @@ def create_interface():
         
         # Improved PDF export functionality
         def export_pdf(report_text, patient_name="Patient", record_id="", age="", gender="", assessment_date="", clinician=""):
+            # Check if ReportLab is available
+            if not REPORTLAB_AVAILABLE:
+                return "Error: ReportLab library is not installed. Please install it with 'pip install reportlab'."
+                
             try:
-                from reportlab.lib.pagesizes import letter
-                from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, Table, TableStyle
-                from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
-                from reportlab.lib import colors
-                import tempfile
-                import webbrowser
-                import os
-                import shutil
                 
                 # Create a proper downloads directory in the app folder
                 downloads_dir = os.path.join(DATA_DIR, "downloads")
@@ -2060,4 +2490,4 @@ if __name__ == "__main__":
     
     # Launch the Gradio app
     app = create_interface()
-    app.launch()
+    app.launch()