NewComer00 · NewComer00 · Mar 22, 2026 · Mar 22, 2026
diff --git a/.gitignore b/.gitignore
@@ -17,7 +17,6 @@ static/vendor/*
 *PitchLoader Output*.ustx
 *output*.ustx
 settings.json
-expressive_config.json
 Expressive-GUI.spec
 .pytest_cache/
 .coverage

diff --git a/README.en.md b/README.en.md
@@ -9,7 +9,7 @@
 
 # Expressive
 
-**Expressive** is a [DiffSinger](https://github.com/openvpi/diffsinger) expression parameter importer developed for [OpenUtau](https://github.com/stakira/OpenUtau). It aims to extract emotional parameters from real human vocals and import them into the appropriate tracks of your project.
+**Expressive** is a [DiffSinger](https://github.com/openvpi/diffsinger) expression parameter importer developed for [OpenUtau](https://github.com/stakira/OpenUtau). It aims to extract expression parameters from real human vocals and import them into the appropriate tracks of your project.
 
 The current version supports importing the following expression parameters:
 
@@ -21,14 +21,14 @@ The current version supports importing the following expression parameters:
   <img src="https://github.com/user-attachments/assets/268b44d4-528d-481e-acfb-3f7da7261c80" width="100%" /> 
 </p>
 
-> - *OpenUtau version used from [keirokeer/OpenUtau-DiffSinger-Lunai](https://github.com/keirokeer/OpenUtau-DiffSinger-Lunai)*
+> - *OpenUtau version from [keirokeer/OpenUtau-DiffSinger-Lunai](https://github.com/keirokeer/OpenUtau-DiffSinger-Lunai)*
 > - *Singer model from [yousa-ling-official-production/yousa-ling-diffsinger-v1](https://github.com/yousa-ling-official-production/yousa-ling-diffsinger-v1)*
 
 > [!TIP]
 > <details>
 >   <summary><b>👉 Click to expand the full voiced demo video 👈</b></summary>
 >
-> https://github.com/user-attachments/assets/4b5b7c15-947a-4f54-b80e-a14a9eefc86b
+>   <p align="center"><video src="https://github.com/user-attachments/assets/4b5b7c15-947a-4f54-b80e-a14a9eefc86b"></video></p>
 >
 > </details>
 
@@ -40,7 +40,7 @@ The current version supports importing the following expression parameters:
 
 By default, this application uses [swift-f0](https://github.com/lars76/swift-f0) (based on ONNX Runtime) as the pitch extraction backend, which runs on CPU only and satisfies basic usage scenarios.
 
-The classic [CREPE](https://github.com/marl/crepe) pitch extraction backend (depends on TensorFlow) is also available, which suits for scenarios with higher accuracy requirements. If your computer is equipped with an NVIDIA GPU and supports [CUDA 11.x](https://docs.nvidia.com/deploy/cuda-compatibility/minor-version-compatibility.html) (i.e., GPU driver version >= 450), the CREPE backend will automatically enable GPU acceleration.
+The classic [CREPE](https://github.com/marl/crepe) pitch extraction backend (depends on TensorFlow) is also available for scenarios with higher accuracy requirements. If your computer is equipped with an NVIDIA GPU and supports [CUDA 11.x](https://docs.nvidia.com/deploy/cuda-compatibility/minor-version-compatibility.html) (i.e., GPU driver version >= 450), the CREPE backend will automatically enable GPU acceleration.
 
 > \* On Windows, TensorFlow 2.10 is the last version that supports GPU acceleration, and Python 3.10 is the highest Python version supported by its `.whl` files.
 
@@ -52,13 +52,17 @@ When using a DiffSinger virtual singer for covers, users often already have an O
 
 ### Inputs
 
+> [!TIP]
+> Starting from `v0.6.0`, this application supports OpenUtau voice tracks with **multiple parts** and **multiple tempos**.
+
 > [!TIP]
 > Starting from `v0.5.0`, users can define a selection region independently within the full audio of both the **Utau vocal** and the **Reference vocal**. The selected audio segment will be used as the final input.
 
-* **Utau vocal**: Emotionless synthesized vocal output from OpenUtau (WAV format). It's recommended to keep `Tempo` and segmentation as close to the reference vocal as possible.
-* **Reference vocal**: Original human vocal recording (WAV format). You can use tools like [UVR](https://github.com/Anjok07/ultimatevocalremovergui) to remove instrumental and reverb.
+* **Utau vocal**: Emotionless synthesized vocal output from OpenUtau (WAV format). It is recommended to keep the segmentation and tempo as close to the **Reference vocal** as possible, as large discrepancies may affect alignment quality.
+* **Reference vocal**: Original human vocal recording (WAV format). You can use tools like [UVR](https://github.com/Anjok07/ultimatevocalremovergui) or [MSST](https://github.com/SUC-DriverOld/MSST-WebUI) to remove instrumentals, harmonies, and reverb.
 * **Input project**: Original OpenUtau project file (USTX format).
 * **Output path**: Where the new processed project file will be saved.
+* **Track number**: The track number in the OpenUtau project where the **Utau vocal** resides (1-based). Expression parameters will be imported into this track.
 
 ### Output
 
@@ -74,10 +78,6 @@ A new USTX file with expression parameters added. The original project will not
 * [x] `Dynamics` generation
 * [x] `Tension` generation
 
-## ⚠️ Known Issues
-
-1. The current version does not support tempo changes within a single track. It’s recommended to use a consistent tempo throughout the project. This limitation will be addressed in future updates.
-
 ## 🚀 Direct Install
 
 You can download pre-compiled executable files directly from the [Releases](https://github.com/NewComer00/expressive/releases) page:
@@ -100,7 +100,6 @@ Includes CUDA runtime libraries. When used on a computer with an NVIDIA GPU (dri
 
 > [!IMPORTANT]
 > This project uses [Git LFS](https://git-lfs.com/) to store large files such as example audio under `examples/`. Please ensure Git LFS is installed on your system before cloning.
-
 ```bash
 git clone https://github.com/NewComer00/expressive.git --depth 1
 cd expressive
@@ -109,7 +108,6 @@ cd expressive
 ### Install the application
 
 Install the package and its dependencies in a virtual environment:
-
 ```bash
 pip install -e ".[gpu,gui]"
 ```
@@ -129,13 +127,11 @@ After installation, you can use the `expressive` and `expressive-gui` entry poin
 ### Command Line Interface (CLI)
 
 Display help:
-
 ```bash
 expressive --help
 ```
 
 Run example in Windows PowerShell:
-
 ```powershell
 expressive `
   --utau_wav "examples/明天会更好/utau.wav" `
@@ -150,7 +146,6 @@ expressive `
 ```
 
 Run example in Linux shell:
-
 ```bash
 expressive \
   --utau_wav "examples/明天会更好/utau.wav" \
@@ -169,20 +164,24 @@ The output project file will be saved to `examples/明天会更好/output.ustx`.
 ### Graphical User Interface (GUI)
 
 Launch in English:
-
 ```bash
 expressive-gui --lang en
 ```
 
 > [!IMPORTANT]
-> Due to framework limitations, the GUI launched via the `expressive-gui` command currently **does not support drag-and-drop**. To use drag-and-drop, please install the GUI [directly](#-direct-install), or run `expressive_gui.py` as a script:
-> 
+> Due to framework limitations, the GUI launched via the `expressive-gui` command currently **does not support drag-and-drop**. To use drag-and-drop, please [install directly](#-direct-install), or run `expressive_gui.py` as a script:
+>
 > ```bash
 > python expressive_gui.py --lang en
 > ```
 
-## 🔬 Algorithm Workflow
+## 📂 Examples
 
+The [`examples/` directory](examples/) contains several sample projects. You can import the `expressive_config.json` file from any example into the GUI to automatically populate all parameters with the preset values.
+
+If you installed the application from the installer, a shortcut named `Expressive-examples` pointing to the examples directory will appear on your desktop after installation — you can import the config files directly from there.
+
+## 🔬 Algorithm Workflow
 ```mermaid
 graph TB;
   ustx_in[/"OpenUtau Project (USTX)"/]
@@ -191,11 +190,14 @@ graph TB;
   refwav-->feat_pitd
   ustx_in-.->|Export|utauwav
   utauwav-->feat_pitd
-  ustx_in-->|Tempo|time_pitd
+
+  ustx_editor["USTX Editor"]
+  ustx_in-->ustx_editor
+  ustx_editor-->|UProject & Time Axis|PitdLoader
 
   subgraph PitdLoader
     direction TB
-    feat_pitd["Features Extraction<br>Pitch & MFCC"]
+    feat_pitd["Features Extraction<br>Pitch & MFCC & RMS"]
 
     time_pitd["Time Alignment<br>FastDTW"]
     feat_pitd-->time_pitd
@@ -232,3 +234,47 @@ graph TB;
     time_tenc-->get_tenc
   end
 ```
+
+## ⚠️ Troubleshooting
+
+### Drag-and-drop does not work on first launch after installation
+
+#### Symptom
+On Windows 10 / 11, after installing the application from the installer for the **first time** (reinstalling after a previous uninstall does not count), the drag-and-drop functionality does not work.
+
+#### Possible Cause
+The [NiceGUI](https://nicegui.io/) framework's support for drag-and-drop in native applications is not yet fully mature. The drag-and-drop feature in this application is currently implemented via the underlying library [pywebview](https://pywebview.flowrl.com/).
+
+#### Solution
+Relaunching the application should restore normal functionality, and this issue will not occur again on the same system afterward.
+
+#### Future Plan
+The NiceGUI framework has begun improving its drag-and-drop support and should resolve this in a future release.
+
+### PITD expression curve is overall too flat
+
+#### Symptom
+The extracted PITD expression curve is too flat, with almost no significant variation overall. Pitch changes in the reference vocal are not reflected in the expression curve.
+
+#### Possible Cause
+The two confidence thresholds in the PITD extractor are set **too high**, causing many pitch changes to be discarded.
+
+#### Solution
+Try lowering both confidence thresholds. In general, the **Utau vocal** is relatively clean, so it is advisable to first adjust the confidence threshold for the **Reference vocal**.
+
+#### Future Plan
+Introduce a better PITD backend (e.g., [RMVPE](https://github.com/Dream-High/RMVPE)). Add visualization of intermediate results.
+
+### PITD expression curve has sudden jumps or spikes at certain positions
+
+#### Symptom
+The PITD expression curve changes too rapidly at certain positions, with very large jumps or spikes that clearly do not match natural vocal behavior.
+
+#### Possible Cause
+The two confidence thresholds in the PITD extractor are set **too low**, causing erroneous detection results to be accepted.
+
+#### Solution
+Try increasing both confidence thresholds. In general, the **Utau vocal** is relatively clean, so it is advisable to first adjust the confidence threshold for the **Reference vocal**.
+
+#### Future Plan
+Introduce a better PITD backend (e.g., [RMVPE](https://github.com/Dream-High/RMVPE)). Add visualization of intermediate results.
diff --git a/README.md b/README.md
@@ -28,8 +28,8 @@
 > <details>
 >   <summary><b>👉 点击展开完整有声演示视频 👈</b></summary>
 >
-> https://github.com/user-attachments/assets/4b5b7c15-947a-4f54-b80e-a14a9eefc86b
->
+>   <p align="center"><video src="https://github.com/user-attachments/assets/4b5b7c15-947a-4f54-b80e-a14a9eefc86b"></video></p>
+> 
 > </details>
 
 ## ✅ 支持平台
@@ -52,13 +52,17 @@
 
 ### 输入
 
+> [!TIP]
+> 从 `v0.6.0` 开始，本应用支持带有**多分段**与**多曲速**的 OpenUtau 人声音轨。
+
 > [!TIP]
 > 从 `v0.5.0` 开始，用户可以分别在**歌姬音声**与**参考人声**的完整音频中划定选区，选区内的音频段落将作为最终输入。
 
-* **歌姬音声**：由 OpenUtau 输出的无表情虚拟歌声音频（WAV 格式）。建议节奏 (`Tempo`) 和分段尽量与参考人声一致。
-* **参考人声**：原始人声录音（WAV 格式），可使用 [UVR](https://github.com/Anjok07/ultimatevocalremovergui) 等工具去除伴奏与混响。
+* **歌姬音声**：由 OpenUtau 输出的无表情虚拟歌声音频（WAV 格式）。建议分段与曲速尽量与**参考人声**相近，若相差过大可能影响对齐效果。
+* **参考人声**：原始人声录音（WAV 格式），可使用 [UVR](https://github.com/Anjok07/ultimatevocalremovergui) 、[MSST](https://github.com/SUC-DriverOld/MSST-WebUI) 等工具去除伴奏、和声与混响。
 * **输入工程**：原始 OpenUtau 工程文件（USTX 格式）。
 * **输出路径**：处理完成后新工程文件的保存位置。
+* **音轨编号**：OpenUtau 工程中**歌姬音声**所在的音轨编号（从 1 开始）。表情参数会被导入到该音轨中。
 
 ### 输出
 
@@ -74,10 +78,6 @@
 * [x] `Dynamics` 参数生成
 * [x] `Tension` 参数生成
 
-## ⚠️ 已知问题
-
-1. 当前版本尚不支持单一轨道中的 `Tempo` 变化，建议工程全程使用统一节奏。该限制将在未来版本中解决。
-
 ## 🚀 直接安装
 
 您可以直接在 [Releases](https://github.com/NewComer00/expressive/releases) 页面下载预编译的可执行文件:
@@ -179,6 +179,12 @@ expressive-gui --lang zh_CN
 > python expressive_gui.py --lang zh_CN
 > ```
 
+## 📂 示例工程
+
+项目的 [`examples/` 目录](examples/)下存放有多个示例。您可以在图形用户界面中导入相应示例的 `expressive_config.json` 配置文件，将预设的参数一键填写到应用中。
+
+若您是从安装包获取的本应用，安装完毕后示例目录的快捷方式 `Expressive-examples` 将出现在您的桌面，您也可以直接导入其中的配置文件。
+
 ## 🔬 算法流程
 ```mermaid
 graph TB;
@@ -188,11 +194,14 @@ graph TB;
   refwav-->feat_pitd
   ustx_in-.->|Export|utauwav
   utauwav-->feat_pitd
-  ustx_in-->|Tempo|time_pitd
+
+  ustx_editor["USTX Editor"]
+  ustx_in-->ustx_editor
+  ustx_editor-->|UProject & Time Axis|PitdLoader
 
   subgraph PitdLoader
     direction TB
-    feat_pitd["Features Extraction<br>Pitch & MFCC"]
+    feat_pitd["Features Extraction<br>Pitch & MFCC & RMS"]
 
     time_pitd["Time Alignment<br>FastDTW"]
     feat_pitd-->time_pitd
@@ -229,3 +238,47 @@ graph TB;
     time_tenc-->get_tenc
   end
 ```
+
+## ⚠️ 常见问题
+
+### 安装后首次运行图形界面，文件拖拽功能无法正常使用
+
+#### 问题现象
+在 Windows 10 / 11 平台下，通过安装包**首次**安装本应用后（先前安装过再卸载不算），应用的文件拖拽功能无法正常使用。
+
+#### 可能原因
+[NiceGUI](https://nicegui.io/) 框架对原生应用的文件拖拽功能支持尚不完善。目前本应用的文件拖拽功能是基于底层库 [pywebview](https://pywebview.flowrl.com/) 实现的。
+
+#### 解决方案
+重新打开应用后应当可以恢复正常，且该系统今后不会再出现此问题。
+
+#### 未来计划
+NiceGUI 框架已经开始着手改进文件拖拽支持，应该在未来的版本中能够解决此问题。
+
+### PITD 表情曲线整体变化过于平缓
+
+#### 问题现象
+提取出的 PITD 表情曲线过于平缓，整体上几乎没有大的起伏，参考人声中的音高变化并没有反映到表情曲线上。
+
+#### 可能原因
+PITD 表情提取器中，两个置信度阈值设置**过高**，许多音高变化没有被采信。
+
+#### 解决方案
+尝试降低两个置信度阈值。一般来说，**歌姬音声**比较纯净，可以先调整**参考人声**的置信度阈值。
+
+#### 未来计划
+引入更好的 PITD 后端（如 [RMVPE](https://github.com/Dream-High/RMVPE)）。添加中间结果的可视化功能。
+
+### PITD 表情曲线在某些位置变化过快，出现跳跃或毛刺
+
+#### 问题现象
+PITD 表情曲线在某些位置变化过快，出现非常大的跳跃或毛刺，明显不符合人声的变化规律。
+
+#### 可能原因
+PITD 表情提取器中，两个置信度阈值设置**过低**，错误的识别结果被采信。
+
+#### 解决方案
+尝试增加两个置信度阈值。一般来说，**歌姬音声**比较纯净，可以先调整**参考人声**的置信度阈值。
+
+#### 未来计划
+引入更好的 PITD 后端（如 [RMVPE](https://github.com/Dream-High/RMVPE)）。添加中间结果的可视化功能。
diff --git a/build/installer.iss b/build/installer.iss
@@ -16,7 +16,7 @@ LZMAUseSeparateProcess=yes
 LZMANumBlockThreads=8
 SetupIconFile=..\assets\icons\app.ico
 PrivilegesRequired=lowest
-PrivilegesRequiredOverridesAllowed=commandline dialog
+PrivilegesRequiredOverridesAllowed=commandline
 
 [Files]
 Source: "..\dist\Expressive-GUI\*"; DestDir: "{app}"; Flags: recursesubdirs createallsubdirs

diff --git a/examples/Прекрасное Далеко/README.md b/examples/Прекрасное Далеко/README.md
@@ -0,0 +1,15 @@
+# Example: Прекрасное Далеко
+
+## Audio
+- **Source:** Прекрасное Далеко / Wonderful Faraway (из к/ф «Гостья из будущего», 1985)
+- **Music:** Евгений Крылатов
+- **Lyrics:** Юрий Энтин
+
+## Voicebank
+- **Source:** [Ashera Lyre DiffSinger V110](https://vocadb.net/Ar/139321)
+- **Voice Provider:** OtomeHime
+- **Illustration:** OtomeHime
+
+## OpenUtau
+- **Phonemizer:** DiffSinger Russian
+- **Tested on version:** OpenUtau LUNAI Edition v0.1.1.0
diff --git a/examples/Прекрасное Далеко/expressive_config.json b/examples/Прекрасное Далеко/expressive_config.json
@@ -0,0 +1,38 @@
+{
+    "utau_wav": "examples/Прекрасное Далеко/utau.wav",
+    "ref_wav": "examples/Прекрасное Далеко/reference.wav",
+    "ustx_input": "examples/Прекрасное Далеко/project.ustx",
+    "ustx_output": "examples/Прекрасное Далеко/output.ustx",
+    "track_number": 1,
+    "ref_start": "0:01.71",
+    "ref_end": "0:22.37",
+    "utau_start": "0:01.75",
+    "utau_end": "0:22.46",
+    "expressions": {
+        "dyn": {
+            "selected": true,
+            "trim_silence": true,
+            "align_radius": 1,
+            "smoothness": 2,
+            "scaler": 1.3
+        },
+        "pitd": {
+            "selected": true,
+            "backend": "crepe",
+            "confidence_utau": 0.8,
+            "confidence_ref": 0.6,
+            "align_radius": 1,
+            "semitone_shift": 0,
+            "smoothness": 4,
+            "scaler": 2.2
+        },
+        "tenc": {
+            "selected": true,
+            "trim_silence": true,
+            "align_radius": 1,
+            "smoothness": 6,
+            "scaler": 1.0,
+            "bias": 10
+        }
+    }
+}