generate_openai_sample.py

#!/usr/bin/env python3
"""
Utility script to generate the OpenAI TTS sample audio file.

This script generates sample_audio_summary_openai.mp3 using the same script
that's used for the Eleven Labs sample, allowing users to compare voice quality
between the two TTS providers in demo mode.

Usage:
    python generate_openai_sample.py

Requirements:
    - OPENAI_API_KEY must be set in .env file
    - OPENAI_TTS_VOICE must be set in .env file (defaults to 'onyx')
"""

import os
import sys
from dotenv import load_dotenv

# Load environment variables
load_dotenv(override=True)

# Import after loading .env
from openai import OpenAI

# Configuration
OPENAI_API_KEY = os.getenv("OPENAI_API_KEY", "").strip()
OPENAI_TTS_VOICE = os.getenv("OPENAI_TTS_VOICE", "onyx").strip()
OUTPUT_FILE = "sample_audio_summary_openai.mp3"

# The same script used for all sample audio files
SAMPLE_AUDIO_SCRIPT = (
    "Here's your spending summary for January 1st, 2014 to January 31st, 2014. "
    "You had 16 expense transactions, totaling 8 thousand 250 dollars and 65 cents. "
    "Your top spending categories were: First, Home Improvement at 5 thousand dollars, "
    "across 2 transactions. Next, Other Expenses at 1 thousand 985 dollars. "
    "Next, Postage and Shipping at 500 dollars. Next, Gasoline/Fuel at 350 dollars. "
    "And finally, Automotive Expenses at 200 dollars. "
    "Home Improvement represents 61% of your total spending. That's your spending overview."
)


def main():
    print("=" * 60)
    print("OpenAI TTS Sample Audio Generator")
    print("=" * 60)
    
    # Check configuration
    if not OPENAI_API_KEY:
        print("❌ Error: OPENAI_API_KEY not found in .env file")
        sys.exit(1)
    
    if not OPENAI_TTS_VOICE:
        print("❌ Error: OPENAI_TTS_VOICE not found in .env file")
        sys.exit(1)
    
    print(f"API Key: {OPENAI_API_KEY[:20]}...")
    print(f"Voice: {OPENAI_TTS_VOICE}")
    print(f"Output: {OUTPUT_FILE}")
    print(f"Script length: {len(SAMPLE_AUDIO_SCRIPT)} characters")
    print()
    
    try:
        print("🔄 Generating audio with OpenAI TTS...")
        client = OpenAI(api_key=OPENAI_API_KEY)
        
        response = client.audio.speech.create(
            model="tts-1",
            voice=OPENAI_TTS_VOICE,
            input=SAMPLE_AUDIO_SCRIPT,
            response_format="mp3"
        )
        
        # Save the audio file
        with open(OUTPUT_FILE, "wb") as f:
            f.write(response.content)
        
        file_size = os.path.getsize(OUTPUT_FILE)
        print()
        print(f"✅ Successfully generated: {OUTPUT_FILE}")
        print(f"   File size: {file_size:,} bytes")
        print()
        print("You can now restart the Gradio app to use provider-specific")
        print("sample audio files in demo mode.")
        
    except Exception as e:
        print(f"❌ Error: {e}")
        sys.exit(1)


if __name__ == "__main__":
    main()