import subprocess
import re
import os

def generate_audio(text, lang, filename):
    subprocess.run(['espeak-ng', '-v', lang, '-s', '120', '-w', filename, text])

def combine_audio(files, output):
    inputs = []
    for file in files:
        inputs.extend(['-i', file])
    filter_complex = ''.join([f'[{i}:a]' for i in range(len(files))]) + f'concat=n={len(files)}:v=0:a=1[out]'
    subprocess.run(['ffmpeg', *inputs, '-filter_complex', filter_complex, '-map', '[out]', output])

def process_text(k, text):
    
    parts = re.split(r'(\[zh\].*?\[/zh\])', text.replace('*',''))
    audio_files = []
    for i, part in enumerate(parts):
        if part.startswith('[zh]') and part.endswith('[/zh]'):
            chinese_text = part[4:-5]
            filename = f'chinese{i}.wav'
            generate_audio(chinese_text, 'cmn', filename)
        else:
            filename = f'english{i}.wav'
            generate_audio(part, 'en', filename)
        audio_files.append(filename)
    num_file=f'explain_{k+1}.wav'
    print(num_file)
    combine_audio(audio_files, num_file )
    
    # Delete temporary audio files
    for file in audio_files:
        os.remove(file)



for i in range(500):
    file_to_open = f'chinese_{i+1}.txt'
    print(file_to_open)
    try: 
      with open(file_to_open, 'r') as file:
          content = file.read()
          print(content)
          process_text(i, content)
    except:
        print('not found')