Skip to content

Commit

Permalink
update readme for release 2.0.3
Browse files Browse the repository at this point in the history
  • Loading branch information
Vladislav committed Jul 5, 2024
1 parent afe4716 commit cc13996
Show file tree
Hide file tree
Showing 38 changed files with 12 additions and 704 deletions.
7 changes: 6 additions & 1 deletion CHANGELOG.json
View file
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
{
"version": "2.0.2",
"version": "2.0.3",
"history": {
"1.4.0": [
"Introduced deepfake synchronization with source video for lip movements",
Expand Down Expand Up @@ -100,6 +100,11 @@
"Added functionality to clone voice in any language from text and audio",
"Improved audio separator to remove noise before clone voice",
"Enhanced processing speed"
],
"2.0.3": [
"Added control restyling (This work with 8 GB VRAM)",
"Added generation text-to-video, image-to-video, text-to-image, outpaint and inpaint image",
"Improved quality of restyling"
]
}
}
218 changes: 2 additions & 216 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@
[![YouTube Chanel](https://img.shields.io/badge/YouTube-Chanel-red)](https://youtube.com/playlist?list=PLJG0sD6007zFJyV78mkU-KW2UxbirgTGr&feature=shared)
[![Hugging Face Models](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Models-yellow)](https://huggingface.co/wladradchenko/wunjo.wladradchenko.ru)

<p align="right">[<a href="README_ru.md">RU v2.0</a>] [see <a href="https://github.com/wladradchenko/wunjo.wladradchenko.ru/tree/v1.6">v1.6</a>]</p>
<p align="right">[<a href="README_ru.md">RU v2.0</a>]</p>
<div id="top"></div>

<br />
Expand Down Expand Up @@ -83,8 +83,6 @@ Wunjo harnesses the power of neural networks to provide cutting-edge solutions i
<details>
<summary><b>Open list to see more details.</b></summary>

In <a href="https://github.com/wladradchenko/wunjo.wladradchenko.ru/releases">version 1.6</a>, Wunjo supports three languages for old approach voice cloning and text to speech synthesis, TTS model training, audio enhancement, a multi-language interface, and the use of the "pass" keyword in the video-to-video feature for Community Edition.

The new version 2.0 introduces a completely redesigned interface, enhanced performance, improved face swap functionality, a generator for non-existent people’s faces, and a deepfake analyzer. It also includes improvements in lip sync, content enhancement, object, text, and background removal, as well as content and component restyling using text prompts. Additionally, public links enable community creation and collaboration.
For audio you will able to use improved audio separator and <b>clone voice on any language from text and audio</b>.

Expand Down Expand Up @@ -125,219 +123,7 @@ For detailed instructions about setup Wunjo CE from GitHub, refer to the [Launch
</div>

<!-- DOWNLOAD -->
You will find on [wunjo.onnline](https://wunjo.online) official installers on Wundows/Ubuntu or portable versions 1.6 and 2.0.

<!-- EXAMPLE -->
## Examples

<b>Open list to see details about requirements and examples.</b>

<details>
<summary>1. Speech synthesis and voice cloning</summary>

Available on CE 1.6.

- [Russian synthesized voice from text](https://soundcloud.com/vladislav-radchenko-234338135/russian-voice-text-synthesis?si=ebfc8ea75d0f4c56a3012ca4fdfb6ab5&utm_source=clipboard&utm_medium=text&utm_campaign=social_sharing)
- [English voice cloned from previously synthesized Russian voice](https://soundcloud.com/vladislav-radchenko-234338135/english-voice-clone?si=057718ee0e714e79b2023ce2e37dfb39&utm_source=clipboard&utm_medium=text&utm_campaign=social_sharing)
- [Chinese voice cloned from a previously synthesized Russian voice](https://soundcloud.com/vladislav-radchenko-234338135/chinese-voice-clone?si=43d437bbdf4d4d9a80c7a4a0031189c0&utm_source=clipboard&utm_medium=text&utm_campaign=social_sharing)

- [Russian audio song](https://soundcloud.com/vladislav-radchenko-234338135/kotletki-s-pyureshkoy-mp3storecc?si=b1f80ace3216458e8f40ea757388b08d&utm_source=clipboard&utm_medium=text&utm_campaign=social_sharing)
- [Russian vocal extract from audio](https://soundcloud.com/vladislav-radchenko-234338135/vocals?si=91b0d2e903204f73b5f3133c0a3a6d7f&utm_source=clipboard&utm_medium=text&utm_campaign=social_sharing)
- [Russian clone voice from audio](https://soundcloud.com/vladislav-radchenko-234338135/russian-clone-voice-from-song-wunjo-ai-161?si=2c4e6929313346eeaeb95da849e8e4d7&utm_source=clipboard&utm_medium=text&utm_campaign=social_sharing)
</details>

<details>
<summary>2. Face animation from image src</summary>

<div align="center">
<table>
<tr>
<th>Original</th>
<th>Fix face + Enhancer</th>
</tr>
<tr align="center">
<td><img src="example/animation_face/samurai_original.gif" alt="original" width="228" height="360"></td>
<td><img src="example/animation_face/samurai_static_enhancer.gif" alt="static_enhancer" width="228" height="360"></td>
</tr>
</table>
</div>
</details>

<details>
<summary>3. Mouth animation from video src</summary>

<div align="center">
<table>
<tr>
<th>Original</th>
<th>Mouth animation</th>
<th>Mouth animation + Enhancer</th>
</tr>
<tr align="center">
<td><img src="example/animate_mouth/pirate_original.gif" alt="original" width="228" height="228"></td>
<td><img src="example/animate_mouth/pirate_mouth.gif" alt="move_enhancer" width="228" height="228"></td>
<td><img src="example/animate_mouth/pirate_mouth_enhancer.gif" alt="static_enhancer" width="228" height="228"></td>
</tr>
</table>
</div>
</details>

<details>
<summary>4. Face swap by one photo</summary>

<div align="center">
<table>
<tr>
<th>Original photo</th>
<th>Original video</th>
<th>Face swap + Background enhancer</th>
</tr>
<tr align="center">
<td><img src="example/face_swap/face_swap_photo_original.gif" alt="original" width="203" height="203"></td>
<td><img src="example/face_swap/face_swap_original.gif" alt="original" width="360" height="203"></td>
<td><img src="example/face_swap/face_swap_smith.gif" alt="static_enhancer" width="360" height="203"></td>
</tr>
</table>
</div>
</details>

<details>
<summary>5. Remove object</summary>

<div align="center">
<table>
<tr>
<th>Original video</th>
<th>Remove object</th>
</tr>
<tr align="center">
<td><img src="example/remove/remove_object_original.gif" alt="original" width="480" height="270"></td>
<td><img src="example/remove/remove_object_retouch.gif" alt="original" width="480" height="270"></td>
</tr>
</table>
</div>
</details>

<details>
<summary>6. Get segmentation mask by one click</summary>

<div align="center">
<table>
<tr>
<th>Original</th>
<th>Mask</th>
</tr>
<tr align="center">
<td><img src="example/remove/segmentation_original.gif" alt="original" width="360" height="360"></td>
<td><img src="example/remove/segmentstion_mask.gif" alt="original" width="360" height="360"></td>
</tr>
</table>
</div>
</details>

<details>
<summary>7. Video-to-Video by Text Prompt (Only for GPU)</summary>

The __higher the video resolution__, __the better the quality__ of the drawn frames.

#### Video resolution 512x512 default model for deepfake

<div align="center">
<table>
<tr>
<th>Original</th>
<th>Blonde hair + Brown jacket</th>
</tr>
<tr align="center">
<td><img src="example/diffusion/diffusion_original.gif" alt="original" width="360" height="360"></td>
<td><img src="example/diffusion/diffusion_object_change.gif" alt="pass" width="360" height="360"></td>
</tr>
</table>
</div>

#### Video resolution 512x512 [custom model for anime](https://civitai.com/models/7371/rev-animated?modelVersionId=19575)

Additionally, you can use your custom stable diffusion model to redraw video or objects in video with difference timeline.

<div align="center">
<table>
<tr>
<th>Original</th>
<th>Object pass + Background change</th>
</tr>
<tr align="center">
<td><img src="example/diffusion/original.gif" alt="original" width="400" height="225"></td>
<td><img src="example/diffusion/pass.gif" alt="pass" width="400" height="225"></td>
</tr>
<tr>
<th>Object change + Enhancement move</th>
<th>Object change + Enhancement anime</th>
</tr>
<tr align="center">
<td><img src="example/diffusion/upscale_move.gif" alt="full_change" width="400" height="225"></td>
<td><img src="example/diffusion/upscale_anime.gif" alt="full_change" width="400" height="225"></td>
</tr>
</table>
</div>

#### Limit resolution video by GPU VRAM

<div align="center">
<table>
<tr>
<th>32 GB</th>
<th>23 GB</th>
<th>18 GB</th>
<th>14 GB</th>
<th>10 GB</th>
<th>8 GB</th>
<th>7 GB</th>
</tr>
<tr align="center">
<th>1280x1280</th>
<th>1080x1080</th>
<th>1024x1024</th>
<th>768x768</th>
<th>640x640</th>
<th>576x576</th>
<th>512x512</th>
</tr>
</table>
</div>
</details>

<details>
<summary>8. Emotion deepfake [Experimental]</summary>

This is an experimental feature that is under development, but you can take a look at some of the work right now in Wunjo CE 1.6.

<div align="center">
<table>
<tr>
<th>Original</th>
<th>Happy</th>
<th>Angry</th>
</tr>
<tr align="center">
<td><img src="example/fake_emotion/original.gif" alt="original" width="213" height="213"></td>
<td><img src="example/fake_emotion/happy.gif" alt="original" width="213" height="213"></td>
<td><img src="example/fake_emotion/angry.gif" alt="original" width="213" height="213"></td>
</tr>
<tr>
<th>Fear</th>
<th>Sad</th>
<th>Disgust</th>
</tr>
<tr align="center">
<td><img src="example/fake_emotion/fear.gif" alt="original" width="213" height="213"></td>
<td><img src="example/fake_emotion/sad.gif" alt="original" width="213" height="213"></td>
<td><img src="example/fake_emotion/disgust.gif" alt="original" width="213" height="213"></td>
</tr>
</table>
</div>
</details>

<!-- EXAMPLE -->
You will find on [wunjo.onnline](https://wunjo.online) official installers on Wundows/Ubuntu.

<!-- UPDATE -->

Expand Down
Loading

0 comments on commit cc13996

Please sign in to comment.