LightSpeech icon indicating copy to clipboard operation
LightSpeech copied to clipboard

LightSpeech: Lightweight and Fast Text to Speech with Neural Architecture Search

Results 5 LightSpeech issues
Sort by recently updated
recently updated
newest added

https://github.com/rishikksh20/LightSpeech/blob/d9290f755f02d33d520c2304c5b6624f87864e55/core/duration_modeling/length_regulator.py#L89 Runtime error when length predicted negative value. Maybe need to modify to: `out.append(x_.repeat(max(int(d_), 0), 1)) `

Hi, thanks for sharing the code. When I tried to run lightspeech on my own dataset, I found in sample file list, the sum duration always starts from -2: https://github.com/rishikksh20/LightSpeech/blob/d9290f755f02d33d520c2304c5b6624f87864e55/filelists/valid_filelist.txt#L1...

Hi, Is it possible to share the pretrained model?

I have 2 question in filelist: 1. Why `-2` is the default start point? 2. Some phonemes are 1 less in length than durations. What does it mean? Thanks for...