From 1d7e8fc637f184e167df199ac37e40c395583152 Mon Sep 17 00:00:00 2001 From: Plachta Date: Mon, 12 Jun 2023 18:26:59 +0800 Subject: [PATCH] Added guidance for training on local machine --- LOCAL.md | 8 ++++++-- README.md | 5 +++-- 2 files changed, 9 insertions(+), 4 deletions(-) diff --git a/LOCAL.md b/LOCAL.md index 416f334..931b1bf 100644 --- a/LOCAL.md +++ b/LOCAL.md @@ -1,6 +1,6 @@ # Train locally ### Build environment -0. Make sure you have Python>=3.6, <=3.8; +0. Make sure you have installed `Python>=3.6, <=3.8`, CMake & C/C++ compilers; 1. Clone this repository; 2. Run `pip install -r requirements`; 3. Install GPU version PyTorch: (Make sure you have CUDA 11.6 or 11.7 installed) @@ -98,4 +98,8 @@ Do replace `{Maximum_epochs}` with your desired number of epochs. Empirically, 100 or more is recommended. To view training progress, open a new terminal and `cd` to the project root directory, run `tensorboard --logdir="./OUTPUT_MODEL"`, then visit `localhost:6006` with your web browser. 12. After training is completed, you can use your model by running: - `python VC_inference.py --model_dir ./OUTPUT_MODEL/G_latest.pth --share True` \ No newline at end of file + `python VC_inference.py --model_dir ./OUTPUT_MODEL/G_latest.pth --share True` +13. To clear all audio data, run: + ``` + rm -rf ./custom_character_voice/* ./video_data/* ./raw_audio/* ./denoised_audio/* ./segmented_character_voice/* long_character_anno.txt short_character_anno.txt + ``` \ No newline at end of file diff --git a/README.md b/README.md index 9e1e198..b7e8f73 100644 --- a/README.md +++ b/README.md @@ -26,8 +26,9 @@ Chinese & Japanese:[![Hugging Face Spaces](https://img.shields.io/badge/%F0%9F ## Fine-tuning -It's recommended to perform fine-tuning on [Google Colab](https://colab.research.google.com/drive/1pn1xnFfdLK63gVXDwV4zCXfVeo8c-I-0?usp=sharing) -because the original VITS has some dependencies that are difficult to configure. +See [LOCAL.md](https://github.com/Plachtaa/VITS-fast-fine-tuning/blob/main/LOCAL.md) for local training guide. +Alternatively, you can perform fine-tuning on [Google Colab](https://colab.research.google.com/drive/1pn1xnFfdLK63gVXDwV4zCXfVeo8c-I-0?usp=sharing) + ### How long does it take? 1. Install dependencies (3 min)