agent/controller.py

from agent.planner import prepare_script
from processing.slide_parser import parse_slides
from utils.translator import translate_text
from speech.tts_engine import generate_voice


def generate_lecture(ppt_file, language, progress=0):

    slides = parse_slides(ppt_file)

    audios = []
    scripts = []

    total = len(slides)

    for idx, slide in enumerate(slides):

        slide_text = slide["text"]

        script = prepare_script(slide_text)

        translated = translate_text(script, language)

        audio = generate_voice(translated, language)

        audios.append(audio)

        scripts.append(f"Slide {slide['index']}:\n{translated}\n")

        progress = int(((idx+1)/total)*100)

    return audios, "\n".join(scripts), progress