Hugging Face's logo

zeroweight-ai
/
ZeroTTS

Text-to-Speech
ONNX
Vietnamese
tts
vietnamese
onnxruntime
zero-shot
speech-synthesis
voice-cloning
vietnamese-tts
tieng-viet
Eval Results (legacy)
Model card Files
xet
 Community
25

Precompute cross-attention K/V in the text encoder

#20
by zeroweightai - opened
base: refs/heads/main
←
from: refs/pr/20
Discussion Files changed
+4
-4
Files changed (2) hide show
  1. onnx/prefix_step.onnx +2 -2
  2. onnx/text_encoder.onnx +2 -2
onnx/prefix_step.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7590e502e32e31cc8afd375a6113c974fd7524efe9d9b9b5b5ed63186607f235
3
- size 391035953
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7544c9fdd3535b21fe8dea855407180ad2084f9da298f8a153893522496565b
3
+ size 348303894
onnx/text_encoder.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b14e0e3df7431c964bcecbb8aa74cf93f7f6daf70cec04cf157ac539a7bf485c
3
- size 280483795
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d37557a4abe07953a02686a176664486e73838389a14ff4a3f36b47ac2bccd52
3
+ size 323096725