| import gradio as gr | |
| import os | |
| import time | |
| from moviepy.editor import * | |
| from share_btn import community_icon_html, loading_icon_html, share_js | |
| token = os.environ.get('HF_TOKEN') | |
| caption = gr.Blocks.load(name="spaces/SRDdev/Image-Caption") | |
| audio_gen = gr.Blocks.load(name="spaces/fffiloni/audioldm-text-to-audio-generation-clone", api_key=token) | |
| ph_message="If you're not happy with sound result, you can manually describe the scene depicted in your image :)" | |
| def infer(image_input, manual_caption, duration_in): | |
| print(duration_in) | |
| if manual_caption == "": | |
| cap = caption(image_input, fn_index=0) | |
| print("gpt2 caption: " + cap) | |
| ph_update = "GP2 Caption: " + cap | |
| else: | |
| cap = manual_caption | |
| print("manual captiony: " + cap) | |
| ph_update="" | |
| sound = audio_gen(cap, duration_in, 2.5, 45, 3, fn_index=0) | |
| return cap, sound[1], gr.Textbox.update(placeholder=f"{ph_update} {ph_message}"), gr.Group.update(visible=True) | |
| title = """ | |
| <div style="text-align: center; max-width: 700px; margin: 0 auto;"> | |
| <div | |
| style=" | |
| display: inline-flex; | |
| align-items: center; | |
| gap: 0.8rem; | |
| font-size: 1.75rem; | |
| " | |
| > | |
| <h1 style="font-weight: 900; margin-bottom: 7px; margin-top: 5px;"> | |
| Image to Sound Effect | |
| </h1> | |
| </div> | |
| <p style="margin-bottom: 10px; font-size: 94%"> | |
| Convert an image to a corresponding sound effect generated through GPT2 Image Captioning & AudioLDM | |
| </p> | |
| </div> | |
| """ | |
| article = """ | |
| <div class="footer"> | |
| <p> | |
| Follow <a href="https://twitter.com/fffiloni" target="_blank">Sylvain Filoni</a> for future updates 🤗 | |
| </p> | |
| </div> | |
| <div id="may-like-container" style="display: flex;justify-content: center;flex-direction: column;align-items: center;margin-bottom: 30px;"> | |
| <p>You may also like: </p> | |
| <div id="may-like-content" style="display:flex;flex-wrap: wrap;align-items:center;height:20px;"> | |
| <svg height="20" width="208" style="margin-left:4px;margin-bottom: 6px;"> | |
| <a href="https://huggingface.co/spaces/haoheliu/audioldm-text-to-audio-generation" target="_blank"> | |
| <image href="https://img.shields.io/badge/🤗 Spaces-AudioLDM_Text_to_Audio-blue" src="https://img.shields.io/badge/🤗 Spaces-AudioLDM_Text_to_Audio-blue.png" height="20"/> | |
| </a> | |
| </svg> | |
| <svg height="20" width="122" style="margin-left:4px;margin-bottom: 6px;"> | |
| <a href="https://huggingface.co/spaces/fffiloni/spectrogram-to-music" target="_blank"> | |
| <image href="https://img.shields.io/badge/🤗 Spaces-Riffusion-blue" src="https://img.shields.io/badge/🤗 Spaces-Riffusion-blue.png" height="20"/> | |
| </a> | |
| </svg> | |
| </div> | |
| </div> | |
| """ | |
| with gr.Blocks(css="style.css") as demo: | |
| with gr.Column(elem_id="col-container"): | |
| gr.HTML(title) | |
| input_img = gr.Image(type="filepath", elem_id="input-img") | |
| manual_cap = gr.Textbox(label="Manual Image description (optional)", lines=2, placeholder=ph_message) | |
| duration_in = gr.Slider(minimum=5, maximum=30, step=5, value=10, label="Duration") | |
| caption_output = gr.Textbox(label="Caption", visible=False, elem_id="text-caption") | |
| sound_output = gr.Audio(label="Result", elem_id="sound-output") | |
| generate = gr.Button("Generate SFX from Image") | |
| with gr.Group(elem_id="share-btn-container", visible=False) as share_group: | |
| community_icon = gr.HTML(community_icon_html) | |
| loading_icon = gr.HTML(loading_icon_html) | |
| share_button = gr.Button("Share to community", elem_id="share-btn") | |
| gr.HTML(article) | |
| generate.click(infer, inputs=[input_img, manual_cap, duration_in], outputs=[caption_output, sound_output, manual_cap, share_group], api_name="i2fx") | |
| share_button.click(None, [], [], _js=share_js) | |
| demo.queue(max_size=32).launch(debug=True) | |