| import os, sys |
| import gradio as gr |
| from huggingface_hub import snapshot_download |
|
|
| |
| |
| from src.gradio_demo import SadTalker |
|
|
| |
|
|
| def toggle_audio_file(choice): |
| """Bascule l'affichage entre l'audio standard et le mode IDLE (pas d'audio requis).""" |
| if choice == False: |
| |
| return gr.update(visible=True), gr.update(visible=False) |
| else: |
| |
| return gr.update(visible=False), gr.update(visible=True) |
| |
| def ref_video_fn(path_of_ref_video): |
| """Met à jour la checkbox 'Use Reference Video' si une vidéo de référence est chargée.""" |
| if path_of_ref_video is not None: |
| return gr.update(value=True) |
| else: |
| return gr.update(value=False) |
| |
| def download_model(): |
| """Télécharge les checkpoints du modèle SadTalker.""" |
| REPO_ID = 'vinthony/SadTalker-V002rc' |
| |
| snapshot_download(repo_id=REPO_ID, local_dir='./checkpoints', local_dir_use_symlinks=False) |
|
|
| |
|
|
| def sadtalker_demo(): |
|
|
| |
| print("Téléchargement des modèles SadTalker...") |
| download_model() |
|
|
| |
| sad_talker = SadTalker(lazy_load=True) |
|
|
| with gr.Blocks(analytics_enabled=False) as sadtalker_interface: |
| gr.Markdown("<div align='center'> <h2> 😭 SadTalker: Learning Realistic 3D Motion Coefficients for Stylized Audio-Driven Single Image Talking Face Animation </span> </h2> \ |
| <a style='font-size:18px;color: #efefef' href='https://arxiv.org/abs/2211.12194'>Arxiv</a> \ |
| <a style='font-size:18px;color: #efefef' href='https://sadtalker.github.io'>Homepage</a> \ |
| <a style='font-size:18px;color: #efefef' href='https://github.com/Winfredy/SadTalker'> Github </div>") |
| |
| |
| gr.Markdown(""" |
| <b>🚀 Dupliquez l'espace et mettez à niveau vers un GPU pour de meilleures performances et une inférence plus rapide sans attendre dans la file d'attente. <a style='display:inline-block' href="https://huggingface.co/spaces/vinthony/SadTalker?duplicate=true"><img src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a></b> \ |
| <br/><b>Alternativement, essayez notre <a href=https://github.com/Winfredy/SadTalker> code GitHub </a> sur votre propre GPU. </b> <a style='display:inline-block' href="https://github.com/Winfredy/SadTalker"><img src="https://img.shields.io/github/stars/Winfredy/SadTalker?style=social"/></a> \ |
| """) |
| |
| with gr.Row(): |
| with gr.Column(variant='panel'): |
| with gr.Tabs(elem_id="sadtalker_source_image"): |
| with gr.TabItem('Source image'): |
| with gr.Row(): |
| source_image = gr.Image(label="Source image", source="upload", type="filepath", elem_id="img2img_image") |
|
|
| with gr.Tabs(elem_id="sadtalker_driven_audio"): |
| with gr.TabItem('Driving Methods'): |
| gr.Markdown("Combinaisons possibles : <br> 1. Audio seul 2. Audio/IDLE Mode + Ref Video 3. IDLE Mode seul 4. Ref Video seul ") |
|
|
| with gr.Row(): |
| driven_audio = gr.Audio(label="Input audio", source="upload", type="filepath", max_length=180) |
| |
| driven_audio_no = gr.Audio(label="Use IDLE mode, no audio is required", source="upload", type="filepath", visible=False) |
|
|
| with gr.Column(): |
| use_idle_mode = gr.Checkbox(label="Use Idle Animation") |
| length_of_audio = gr.Number(value=5, label="The length(seconds) of the generated video.") |
| |
| use_idle_mode.change(toggle_audio_file, inputs=use_idle_mode, outputs=[driven_audio, driven_audio_no]) |
|
|
| with gr.Row(): |
| ref_video = gr.Video(label="Reference Video", source="upload", type="filepath", elem_id="vidref") |
|
|
| with gr.Column(): |
| use_ref_video = gr.Checkbox(label="Use Reference Video") |
| ref_info = gr.Radio(['pose', 'blink','pose+blink', 'all'], value='pose', label='Reference Video',info="Comment emprunter de la vidéo de référence ?") |
|
|
| ref_video.change(ref_video_fn, inputs=ref_video, outputs=[use_ref_video]) |
|
|
|
|
| with gr.Column(variant='panel'): |
| with gr.Tabs(elem_id="sadtalker_checkbox"): |
| with gr.TabItem('Settings'): |
| gr.Markdown("Aide ? Visitez notre [[page de meilleures pratiques](https://github.com/OpenTalker/SadTalker/blob/main/docs/best_practice.md)]") |
| with gr.Column(variant='panel'): |
| |
| with gr.Row(): |
| pose_style = gr.Slider(minimum=0, maximum=45, step=1, label="Pose style", value=0) |
| exp_weight = gr.Slider(minimum=0, maximum=3, step=0.1, label="expression scale", value=1) |
| blink_every = gr.Checkbox(label="use eye blink", value=True) |
|
|
| with gr.Row(): |
| size_of_image = gr.Radio([256, 512], value=256, label='face model resolution', info="Utiliser le modèle 256/512 ?") |
| preprocess_type = gr.Radio(['crop', 'resize','full', 'extcrop', 'extfull'], value='crop', label='preprocess', info="Comment traiter l'image source?") |
| |
| with gr.Row(): |
| is_still_mode = gr.Checkbox(label="Still Mode (moins de mouvement de tête, fonctionne avec `full`)") |
| facerender = gr.Radio(['facevid2vid','pirender'], value='facevid2vid', label='facerender', info="Quel moteur de rendu de visage ?") |
| |
| with gr.Row(): |
| batch_size = gr.Slider(label="batch size in generation", step=1, maximum=10, value=1) |
| enhancer = gr.Checkbox(label="GFPGAN as Face enhancer") |
| |
| submit = gr.Button('Generate', elem_id="sadtalker_generate", variant='primary') |
| |
| with gr.Tabs(elem_id="sadtalker_genearted"): |
| gen_video = gr.Video(label="Generated video", format="mp4", scale=1) |
|
|
| |
| submit.click( |
| fn=sad_talker.test, |
| inputs=[source_image, |
| driven_audio, |
| preprocess_type, |
| is_still_mode, |
| enhancer, |
| batch_size, |
| size_of_image, |
| pose_style, |
| facerender, |
| exp_weight, |
| use_ref_video, |
| ref_video, |
| ref_info, |
| use_idle_mode, |
| length_of_audio, |
| blink_every |
| ], |
| outputs=[gen_video], |
| ) |
|
|
| |
| with gr.Row(): |
| examples = [ |
| [ |
| 'examples/source_image/full_body_1.png', |
| 'examples/driven_audio/bus_chinese.wav', |
| 'crop', |
| True, |
| False |
| ], |
| |
| [ |
| 'examples/source_image/art_5.png', |
| 'examples/driven_audio/RD_Radio31_000.wav', |
| 'resize', |
| True, |
| True |
| ], |
| ] |
| gr.Examples(examples=examples, |
| inputs=[ |
| source_image, |
| driven_audio, |
| preprocess_type, |
| is_still_mode, |
| enhancer], |
| outputs=[gen_video], |
| fn=sad_talker.test, |
| |
| cache_examples=False) |
|
|
| return sadtalker_interface |
| |
|
|
| if __name__ == "__main__": |
|
|
| demo = sadtalker_demo() |
| |
| |
| demo.queue(max_size=10, api_open=True) |
| demo.launch(debug=True) |