 Lengyue
|
362a1a7116
Update dual ar generate & config
|
2 سال پیش |
|
Lengyue
|
9e1a9debfd
fix bug & implement generate
|
2 سال پیش |
|
Lengyue
|
55ac942c83
Optimize llama sft config
|
2 سال پیش |
|
Lengyue
|
0855b454e4
Add VITS vq
|
2 سال پیش |
|
Lengyue
|
37f4b517c2
Implement delay pattern generate
|
2 سال پیش |
|
Lengyue
|
ab91950b9a
Fix bug
|
2 سال پیش |
|
Lengyue
|
a588f458f1
Fix torch compile
|
2 سال پیش |
|
Lengyue
|
a0bf975f55
Add workaround for english no-g2p decoding
|
2 سال پیش |
|
Lengyue
|
bbb492d7eb
Allow disable interactive mode
|
2 سال پیش |
|
Lengyue
|
3a4490c632
Fix codebook inconsistent
|
2 سال پیش |
|
Lengyue
|
5f9e2f2380
Update new config and generation scripts
|
2 سال پیش |
|
Lengyue
|
65e646ae31
Support prompt auto truncating codebook
|
2 سال پیش |
|
Lengyue
|
83582d1e89
Update dataloader, loss tracker, and config
|
2 سال پیش |
|
Lengyue
|
467353b23a
Fix error handling & optimize dataset
|
2 سال پیش |
|
Lengyue
|
982f78794e
Fix lora inference
|
2 سال پیش |
|
Lengyue
|
e7e7a8c2e3
Optimize inifinity generation
|
2 سال پیش |
|
Lengyue
|
3a08434bfb
Add g2p config and semantic convert to api server
|
2 سال پیش |
|
Lengyue
|
0dafcf373d
Reuse cache for different generation size
|
2 سال پیش |
 spicysama
|
06a35aef53
add fastapi for inference (#12)
|
2 سال پیش |
|
Lengyue
|
c583555995
Add half precision inference and document
|
2 سال پیش |
|
Lengyue
|
45b8ed0bb0
Fix inference error
|
2 سال پیش |
|
Lengyue
|
2966ba9019
Support speaker finetune
|
2 سال پیش |
|
Lengyue
|
9e664473b9
Update document
|
2 سال پیش |
|
Lengyue
|
05c342704b
Optimize auto clean text
|
2 سال پیش |
|
Lengyue
|
7ae872018f
Add arg to disable g2p
|
2 سال پیش |
|
Lengyue
|
c54eecc63f
Implement windowed repeat penalty
|
2 سال پیش |
|
Lengyue
|
cb482fee94
Fix primary codebook penalty
|
2 سال پیش |
|
Lengyue
|
3f7b8215bc
Better param & instruction
|
2 سال پیش |
|
Lengyue
|
7a0086536c
Polish inference script
|
2 سال پیش |
|
Lengyue
|
afc854ad85
Clean unused code & reorganize
|
2 سال پیش |
