mirror of
https://github.com/storytold/storyteller-ml.git
synced 2026-10-15 00:03:06 +00:00
Paul Sajna
f86b44d8fc
* 48KHz softvc * add some notes * almost forgot the inference script * fix readme link * requirements.txt
16 lines
1003 B
Markdown
16 lines
1003 B
Markdown
# 48KHz SoftVC
|
|
|
|
Herein lies 48KHz softvc, a failed experiment to produce a high samplerate version of https://github.com/bshall/soft-vc
|
|
The goal was to leverage the strong character quality that softvc has, ie it's ability to work well with weird voices
|
|
such as Evanescence, GladOS, Darth Vader, and Yoda, while providing a higher samplerate than the 16KHz of the original
|
|
version. We used the same technique as so-vits-svc, to upscale the 16KHz HuBERT "speech units" to 48KHz equivalents,
|
|
and trained on 48KHz and 44.1 KHz source data. Unfortunately, the results have a robotic, metallic sound to them,
|
|
and sound basically the same as if it was 16KHz (with weird robotic noises added for flavour).
|
|
|
|
|
|
Additionally, warm-starting didn't seem to work, maybe there's some trick to remove the speaker embedding of the base
|
|
model, but it's not part of the code provided by bshall.
|
|
|
|
|
|
To hear the results for yourself, visit https://drive.google.com/drive/folders/1byrr6R70qr4kUVj39KXi9NCXmbJVJgNU?usp=sharing
|