Built-in pretrained encoder/vocoder model 简化配置流程,预集成模型

.gitignore

@@ -15,6 +15,4 @@
 *.toc
 *.wav
 *.sh
-encoder/saved_models/*
+synthesizer/saved_models/*
-synthesizer/saved_models/*
-vocoder/saved_models/*

README-CN.md

@@ -25,16 +25,7 @@
 * 安装 [ffmpeg](https://ffmpeg.org/download.html#get-packages)。
 * 运行`pip install -r requirements.txt` 来安装剩余的必要包。
 
-### 2. 使用预训练好的编码器/声码器
+### 2. 使用数据集训练合成器
-下载[预训练模型](https://github.com/CorentinJ/Real-Time-Voice-Cloning/wiki/Pretrained-models)，解压下载内容，并复制`encoder`与`vocoder`下的`saved_models`到本代码库的相应目录下
-
-确保得到以下文件：
-```
-encoder\saved_models\pretrained.pt
-vocoder\saved_models\pretrained\pretrained.pt
-```
-
-### 3. 使用数据集训练合成器
 * 下载 数据集并解压：确保您可以访问 *train* 文件夹中的所有音频文件（如.wav）
 * 使用音频和梅尔频谱图进行预处理：
 `python synthesizer_preprocess_audio.py <datasets_root>`
@@ -50,14 +41,15 @@ vocoder\saved_models\pretrained\pretrained.pt
 * 当您在训练文件夹 *synthesizer/saved_models/* 中看到注意线显示和损失满足您的需要时，请转到下一步。
 > 仅供参考，我的注意力是在 18k 步之后出现的，并且在 50k 步之后损失变得低于 0.4。
 
-
+### 3. 启动工具箱
-### 4. 启动工具箱
 然后您可以尝试使用工具箱：
 `python demo_toolbox.py -d <datasets_root>`
 
 ## TODO
-- [ ] 允许直接使用中文
+- [X] 允许直接使用中文
 - [X] 添加演示视频
 - [X] 添加对更多数据集的支持
-- [ ] 上传预训练模型
+- [X] 上传预训练模型
+- [ ] 支持parallel tacotron
+- [ ] 服务化与容器化
 - [ ] 🙏 欢迎补充

README.md

@@ -26,12 +26,8 @@
 * Install [PyTorch](https://pytorch.org/get-started/locally/).
 * Install [ffmpeg](https://ffmpeg.org/download.html#get-packages).
 * Run `pip install -r requirements.txt` to install the remaining necessary packages.
-
+> Note that we are using the pretrained encoder/vocoder but synthesizer, since the original model is incompatible with the Chinese sympols. It means the demo_cli is not working at this moment.
-### 2. Reuse the pretrained encoder/vocoder
+### 2. Train synthesizer with your dataset
-* Download the following models and extract the encoder and vocoder models to the according directory of this project. Don't use the synthesizer
-https://github.com/CorentinJ/Real-Time-Voice-Cloning/wiki/Pretrained-models
-> Note that we need to specify the newly trained synthesizer model, since the original model is incompatible with the Chinese sympols. It means the demo_cli is not working at this moment.
-### 3. Train synthesizer with your dataset
 * Download aidatatang_200zh or SLR68 dataset and unzip: make sure you can access all .wav in *train* folder
 * Preprocess with the audios and the mel spectrograms:
 `python synthesizer_preprocess_audio.py <datasets_root>`
@@ -48,7 +44,7 @@ Allow parameter `--dataset {dataset}` to support adatatang_200zh, SLR68
 ![step-135500-mel-spectrogram_sample_1](https://user-images.githubusercontent.com/7423248/128587255-4945faa0-5517-46ea-b173-928eff999330.png)
 > A link to my early trained model: [Baidu Yun](https://pan.baidu.com/s/10t3XycWiNIg5dN5E_bMORQ)
 Code：aid4
-### 4. Launch the Toolbox
+### 3. Launch the Toolbox
 You can then try the toolbox:
 
 `python demo_toolbox.py -d <datasets_root>`  
@@ -59,4 +55,6 @@ or
 - [x] Add demo video
 - [X] Add support for more dataset
 - [X] Upload pretrained model
+- [ ] Support parallel tacotron
+- [ ] Service orianted and docterize
 - 🙏 Welcome to add more