utils/tts_engine.py

"""
Text-to-Speech Engine
=====================
Converts text to speech audio using Google Text-to-Speech (gTTS).

Functions:
    - synthesize_speech: Convert text to MP3 audio file
    - get_audio_duration: Get duration of audio file
"""

from gtts import gTTS
from pathlib import Path
import time
from typing import Literal
import os


def synthesize_speech(
    text: str, 
    language: Literal["en", "hi"],
    output_dir: Path,
    slow: bool = False
) -> str:
    """
    Convert text to speech and save as MP3 file.
    
    Uses Google Text-to-Speech (gTTS) for natural-sounding
    speech synthesis in English and Hindi.
    
    Args:
        text: Text to convert to speech
        language: Language code ('en' for English, 'hi' for Hindi)
        output_dir: Directory to save the audio file
        slow: If True, speak slowly (useful for language learning)
        
    Returns:
        Path to the generated MP3 file
    """
    # Ensure output directory exists
    output_dir = Path(output_dir)
    output_dir.mkdir(parents=True, exist_ok=True)
    
    # Generate unique filename using timestamp
    timestamp = int(time.time() * 1000)
    output_path = output_dir / f"tts_{timestamp}.mp3"
    
    try:
        # Create TTS object and save to file
        tts = gTTS(text=text, lang=language, slow=slow)
        tts.save(str(output_path))
        
        # Verify file was created
        if not output_path.exists():
            raise FileNotFoundError(f"TTS file was not created: {output_path}")
        
        # Verify file has content
        if output_path.stat().st_size == 0:
            raise ValueError("TTS file is empty")
            
        return str(output_path)
        
    except Exception as e:
        print(f"TTS Error: {e}")
        raise


def get_audio_duration(audio_path: str) -> float:
    """
    Get the duration of an audio file in seconds.
    
    Args:
        audio_path: Path to the audio file
        
    Returns:
        Duration in seconds (estimated if pydub fails)
    """
    try:
        from pydub import AudioSegment
        audio = AudioSegment.from_file(audio_path)
        return len(audio) / 1000.0  # Convert milliseconds to seconds
    except Exception:
        # Fallback: estimate based on file size
        # Approximate: MP3 at 128kbps = 16KB per second
        try:
            file_size = os.path.getsize(audio_path)
            return file_size / 16000
        except Exception:
            return 3.0  # Default 3 seconds