-
I've also implemented an E2E system using a CFM prior (different flow matching architecture instead of 1D-Unet). Despite using the prior loss in Grad-TTS, alignment framework fails to converge. (Prior…
-
Thanks to the author for his/her fantastic model that nearly eliminated the timbre leaking problem.
For someone who doesn't want to use the FairSeq, I made a HuggingFace version of `content vec bes…
-
有些代码它影响我导出model成 torch_script 就是 torch.jit.script 这种。但是他们在上下文中并没有被使用，所以我想问问能不能删掉
我正在导出 vits 和 t2s 成 torch_script 然后在 rust 里面推理。目前可以推理出声音，但是我改动了部分python代码。所以我想问问。
我的改动在这里 https://github.com/L-jasmin…
-
-
- https://openreview.net/forum?id=TVHS5Y4dNvM
- 2021 ICLR 2022 Conference Blind Submission
概要：
視覚タスクにおいては，長年にわたり畳み込みネットワークが主流のアーキテクチャであったが，最近の実験では，Vision Transformer（ViT）に代表されるTransformerベースのモデ…
e4exp updated
3 years ago
-
i finetuned model use custom data, the finetune code:
`#!/bin/bash
now=$(date +"%Y%m%d_%H%M%S")
epoch=120
bs=2
gpus=2
lr=0.000005
encoder=vits
dataset=custom # vkitti
img_size=518
min_de…
-
### Describe the bug
Error while training:-
- I tried with sudo same error
- I am using docker image nvidia/cuda:11.7.0-base-ubuntu22.04
- The default value of the docker container for command `…
-
I am wanting to try out XPhoneBERT for my school project, but the main page [README.md](https://github.com/VinAIResearch/XPhoneBERT/blob/main/README.md) file doesn't clearly show how to generate audio…
-
This issue relates to how we embed each (16x16) patch. Additionally, we discuss the positional encodings we add to each patch's embedding.
# Patch Embedding
Let's review, we split the images in…
-
When I execute the “python dynamo.py export --encoder vits --output weights/vitb.onnx --opset 18” statement, there is an error， “ImportError: cannot import name 'StrEnum' from 'enum' ”，Could it be a p…