Short Chinese sentences seem broken (简短的中文句子似乎不完整)

[ericfarng]

When infer() for short chinese sentence, ”他多高“，”你看看“，”五万“ the audio wavs are often cut off

do not infer() in batch. only infer() a single sentence. I think they are broken because when the audio works with windows media player, but they do not load in VLC media player.

google translate:

当 infer() 短句中文时，”他多高“，”你看看“，”五万“ 音频 wav 经常被切断

不要批量 infer()。只 infer() 单个句子。 我认为它们坏了，因为音频在 Windows Media Player 中可以播放，但在 VLC Media Player 中无法加载。

[ericfarng]

looks like adding [uv_break] seems to help. more details here https://github.com/2noise/ChatTTS/issues/503#issuecomment-2453092580