This is a piece of text-to-speech I have done using XTTS_v2: you can find it here. The sentence is “It brings Vedanta and other spiritual philosophies to common man.”.

My configuration for the XTTS was:

def process_sentence(self, sentence, idx):
    print(f"Processing index {idx}...", flush=True)
    audio_file = f"temp_{idx}.wav"
    self.model.tts_to_file(text=sentence,
    file_path=audio_file,
    language=self.language,
    speed=0.9,
    speaker="Ana Florence",
    emotion="Happy")

Can anybody help me out with the crack in sound when the speaker says “Vedanta”? How can I fix it? I am a noob to TTS and I was hoping to get some help regarding this.