app.py

import os
from PyPDF2 import PdfReader
from gtts import gTTS
from pydub import AudioSegment
import gradio as gr

def pdf_to_text(pdf_file):
    try:
        reader = PdfReader(pdf_file)
        text = ""
        for page in reader.pages:
            text += page.extract_text()
        return text
    except Exception as e:
        return f"Error: {str(e)}"

def text_to_speech(text, lang='en', file_format='mp3'):
    tts = gTTS(text=text, lang=lang)
    output_path = f"output.{file_format}"
    tts.save("temp.mp3")
    
    if file_format == 'mp3':
        os.rename("temp.mp3", output_path)
    elif file_format == 'wav':
        sound = AudioSegment.from_mp3("temp.mp3")
        sound.export(output_path, format="wav")
        os.remove("temp.mp3")
    
    return output_path

def convert_pdf_to_speech(pdf_file, lang='en', file_format='mp3'):
    text = pdf_to_text(pdf_file)
    if text and not text.startswith("Error:"):
        audio_file = text_to_speech(text, lang=lang, file_format=file_format)
        return audio_file
    else:
        return text

# Gradio arayüzü
def interface(pdf_file, lang, file_format):
    audio_file = convert_pdf_to_speech(pdf_file.name, lang=lang, file_format=file_format)
    return audio_file

iface = gr.Interface(
    fn=interface, 
    inputs=[
        gr.File(label="Pdf i yükleyiniz"),
        gr.Dropdown(choices=["en", "es", "fr", "de", "it", "tr", "zh", "jp"], label="Dil Seçiniz"),
        gr.Radio(choices=["mp3", "wav"], label="Çıktı formatını seçiniz")
    ], 
    outputs="file", 
    title="Pdf Seslendirme",
    description="Bir PDF dosyası yükleyin, dili seçin ve çıktı formatını (MP3 veya WAV) seçiniz."
)
iface.launch(share=True)