text to speech with pause duration

dictation.py

from gtts import gTTS #updated 2024-06-19-081215
from pydub import AudioSegment
import os

def text_to_speech_with_pauses(text, lang='zh-CN', pause_duration=8000):
    segments = text.split(',')  # Split text at comma
    combined = AudioSegment.silent(duration=0)  # Start with an empty segment

    for segment in segments:
        tts = gTTS(text=segment, lang=lang)
        temp_file = "temp.mp3"
        tts.save(temp_file)
        
        audio_segment = AudioSegment.from_file(temp_file, format="mp3")
        combined += audio_segment + AudioSegment.silent(duration=pause_duration)
        
        os.remove(temp_file)  # Clean up temporary file

    output_file = "output_with_pauses.mp3"
    combined.export(output_file, format="mp3")
    print(f"Output file saved as {output_file}")

if __name__ == "__main__":
    text = '台湾岛,民族,情谊,奋发,节日,春节,花灯,清明节,先人,龙舟,中秋,转眼,团圆,热闹,动物,贝壳,自己,身体,甲骨文,样子,可以,钱币,钱财,有关,比如,美食,红烧,茄子,烤鸭,羊肉,蛋炒饭,课文,彩色,脚尖,森林,雪松,歌声,苹果,精灵,季节,好像,一直,说话,童话,称呼,对岸,发现,弟弟,游戏,发明,字母,周围,补充,公主,伙伴,飞机,地道,火药,合力,忘记,屁股'
    text_to_speech_with_pauses(text)