The first words are missing from playback

[A-n-d-r-e-a-s-M]

Self Checks

• This template is only for bug reports. For questions, please visit Discussions.
• I have thoroughly reviewed the project documentation (installation, training, inference) but couldn't find information to solve my problem. English 中文 日本語 Portuguese (Brazil)
• I have searched for existing issues, including closed ones. Search issues
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）
• Please do not modify this template and fill in all required fields.

Cloud or Self Hosted

Self Hosted (Source)

Environment Details

WIndows 11, Python 3.10 ,torch 2.4.1.+cu124, torchaudio 2.4.1.+cu124, torchvision 0.19.1++cu124, gradio 5.15.0, test at gradio web ui

Steps to Reproduce

1. I loaded a German 12s audio file and added the correct transcription.
2. I entered the following 2 texts for the synthesis: "Das ist ein schöner Text." and "Hier ist ein schöner Text.".

✔️ Expected Behavior

Complete synthesis of all words.

❌ Actual Behavior

The first one or two words were not or only partially reproduced. This seems to depend on the audio file and does not happen always. But, this file isn't corrupted. Per example with other tts libraries, there are no problems.

[viniciusrtf]

I am seeing gaps in some words in the beginning of the sentence, regardless of if I use a reference audio or not, so this is not an issue of a corrupted reference audio.

[github-actions]

This issue is stale because it has been open for 30 days with no activity.

[github-actions]

This issue was closed because it has been inactive for 14 days since being marked as stale.