import os from gtts import gTTS from app.agent.graph import bfsi_agent from dotenv import load_dotenv load_dotenv() def create_mock_customer_audio(text, lang, filename): print(f"Simulating customer speaking {lang}...") tts = gTTS(text=text, lang=lang) tts.save(filename) return filename def run_test(audio_path, test_name): print(f"\nRunning BFSI Agent Pipeline: {test_name}") initial_state = { "audio_path": audio_path, "customer_id": "C1024", "messages": [] } result = bfsi_agent.invoke(initial_state) # Safely print without emojis to avoid Windows cp1252 codec errors print(f"STT Transcript: {result.get('transcript')}") print(f"Language Detected: {result.get('language')}") print(f"Intent Detected: {result.get('intent')}") print(f"Agent Response: {result.get('response_text')}") print(f"Agent Audio File Generated: {result.get('response_audio_path')}") print("-" * 50) if __name__ == "__main__": print("Starting Multilingual Agent Evaluation...\n") # 1. Hindi Test (Tool Calling Intent) hi_audio = create_mock_customer_audio("मेरे लोन का स्टेटस क्या है?", "hi", "mock_hindi.mp3") run_test(hi_audio, "Hindi Loan Status Query") # 2. Tamil Test (RAG Knowledge Intent) ta_audio = create_mock_customer_audio("தனிநபர் கடனுக்கான வட்டி விகிதம் என்ன?", "ta", "mock_tamil.mp3") run_test(ta_audio, "Tamil Interest Rate Query") print("Multilingual tests completed successfully!")