import gradio as gr caption = gr.Blocks.load(name="spaces/SRDdev/Image-Caption") audio_gen = gr.Blocks.load(name="spaces/haoheliu/audioldm-text-to-audio-generation") def infer(image_input): cap = caption(image_input, fn_index=0) sound = audio_gen(cap, 5, 2.5, 45, 3, fn_index=0) return sound, gr.Group.update(visible=True) title = """

Image to Sound Effect

Convert an image to corresponding sound effect through AudioLDM

""" article = """ """ 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") sound_output = gr.Video(label="Result", elem_id="sound-output") generate = gr.Button("Generate FX 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], outputs=[sound_output, share_group], api_name="i2fx") demo.queue(max_size=32, concurrency_count=20).launch()