The generated output contains repeated words or includes audio from the prompt that ends up in the generated audio

The sentence I used for inference is: The aroma of freshly baked bread filled the kitchen, making everyone's mouth water in [anticipation.](url) The audio output contains repeated words: https://drive.google.com/file/d/15qPUHbfQzZhTjd4vjA_UsedoF-5_z-k5/view?usp=sharing https://drive.google.com/file/d/11dgpqmyhZu36I1du6YmxL6h0OHLKGET1/view?usp=sharing

contains repeated words and includes audio from the prompt: Audio Prompt: https://drive.google.com/file/d/1ybpwClOFV3mUxgfUdYE3ib8Cq3Fe-XPT/view?usp=sharing Audio Output: https://drive.google.com/file/d/1LjoasRXyciPvCH8eiHss1YhZ2chykvUW/view?usp=sharing