fluency/main.py

import json
import whisper
from .compute_fluency import compute_fluency_score

def main():
    """
    Main function to run fluency analysis on audio files
    """
    # Fixed parameters - modify these values directly in the code
    audio_file = r"D:\Intern\shankh\audio_samples\obama_short.wav"  # Path to your audio file
    model_size = "base"  # Whisper model size (tiny, base, small, medium, large)
    verbose = True  # Whether to print detailed results
    
    try:
        # Load whisper model
        print(f"Loading Whisper model ({model_size})...")
        whisper_model = whisper.load_model(model_size)
        
        # Calculate fluency score
        print(f"Analyzing fluency for {audio_file}...")
        results = compute_fluency_score(audio_file, whisper_model)
        
        # Print summary results
        print("\nFluency Analysis Results:")
        print(f"- Fluency Score: {results['fluency_score']:.2f}/100")
        print(f"- Insight: {results['insight']}")
        print(f"- Speech Rate Stability (SRS): {results['SRS']:.2f}/100")
        print(f"- Pause Appropriateness (PAS): {results['PAS']:.2f}/100")
        
        # Print verbose results if enabled
        if verbose:
            print("\nDetailed Metrics:")
            print(f"- Words per minute: {results['components']['wpm']:.1f}")
            print(f"- Filler word count: {results['components']['filler_count']}")
            print(f"- Long pauses: {results['components']['long_pause_count']}")
            print(f"- Pitch variation: {results['components']['pitch_variation']:.2f} semitones")
            print(f"- Natural Pause Placement: {results['components']['pas_components']['NPP']:.2f}/100")
            print(f"- Avoidance of Filler Words: {results['components']['pas_components']['AFW']:.2f}/100")
            
            # Print first 100 characters of transcript
            transcript_preview = results['transcript'][:] + "..." if len(results['transcript']) > 100 else results['transcript']
            print(f"\nTranscript preview: {transcript_preview}")
        
    except Exception as e:
        print(f"Error during analysis: {str(e)}")
        return 1

if __name__ == "__main__":
    exit(main())