app.py

# Adapted app.py from https://huggingface.co/spaces/haoheliu/audioldm-text-to-audio-generation/blob/main/app.py
import gradio as gr
import torch
from diffusers import AudioLDMPipeline

from transformers import AutoProcessor, ClapModel

# replace with cuda code from AudioLDM's original app.py if using GPU
device = "cpu"
torch_dtype = torch.float32

# load AudioLDM Diffuser Pipeline
pipe = AudioLDMPipeline.from_pretrained("cvssp/audioldm-m-full", torch_dtype=torch_dtype).to(device)
pipe.unet = torch.compile(pipe.unet)

# omit CLAP model because we'll only generate one waveform, no scoring

generator = torch.Generator(device)

# modified from audioldm app.py to omit n_candidates
def text2audio(text, negative_prompt, duration, guidance_scale, random_seed):
    if text is None:
        raise gr.Error("Please provide a text input.")

    waveforms = pipe(
        text,
        audio_length_in_s=duration,
        guidance_scale=guidance_scale,
        negative_prompt=negative_prompt,
        num_waveforms_per_prompt=1,
        generator=generator.manual_seed(int(random_seed)),
    )["audios"]

    waveform = waveforms[0]

    return gr.make_waveform((16000, waveform), bg_image="bg.png")

# duplicate CSS config

css = """
        a {
            color: inherit; text-decoration: underline;
        } .gradio-container {
            font-family: 'IBM Plex Sans', sans-serif;
        } .gr-button {
            color: white; border-color: #000000; background: #000000;
        } input[type='range'] {
            accent-color: #000000;
        } .dark input[type='range'] {
            accent-color: #dfdfdf;
        } .container {
            max-width: 730px; margin: auto; padding-top: 1.5rem;
        } #gallery {
            min-height: 22rem; margin-bottom: 15px; margin-left: auto; margin-right: auto; border-bottom-right-radius:
            .5rem !important; border-bottom-left-radius: .5rem !important;
        } #gallery>div>.h-full {
            min-height: 20rem;
        } .details:hover {
            text-decoration: underline;
        } .gr-button {
            white-space: nowrap;
        } .gr-button:focus {
            border-color: rgb(147 197 253 / var(--tw-border-opacity)); outline: none; box-shadow:
            var(--tw-ring-offset-shadow), var(--tw-ring-shadow), var(--tw-shadow, 0 0 #0000); --tw-border-opacity: 1;
            --tw-ring-offset-shadow: var(--tw-ring-inset) 0 0 0 var(--tw-ring-offset-width)
            var(--tw-ring-offset-color); --tw-ring-shadow: var(--tw-ring-inset) 0 0 0 calc(3px
            var(--tw-ring-offset-width)) var(--tw-ring-color); --tw-ring-color: rgb(191 219 254 /
            var(--tw-ring-opacity)); --tw-ring-opacity: .5;
        } #advanced-btn {
            font-size: .7rem !important; line-height: 19px; margin-top: 12px; margin-bottom: 12px; padding: 2px 8px;
            border-radius: 14px !important;
        } #advanced-options {
            margin-bottom: 20px;
        } .footer {
            margin-bottom: 45px; margin-top: 35px; text-align: center; border-bottom: 1px solid #e5e5e5;
        } .footer>p {
            font-size: .8rem; display: inline-block; padding: 0 10px; transform: translateY(10px); background: white;
        } .dark .footer {
            border-color: #303030;
        } .dark .footer>p {
            background: #0b0f19;
        } .acknowledgments h4{
            margin: 1.25em 0 .25em 0; font-weight: bold; font-size: 115%;
        } #container-advanced-btns{
            display: flex; flex-wrap: wrap; justify-content: space-between; align-items: center;
        } .animate-spin {
            animation: spin 1s linear infinite;
        } @keyframes spin {
            from {
                transform: rotate(0deg);
            } to {
                transform: rotate(360deg);
            }
        } #share-btn-container {
            display: flex; padding-left: 0.5rem !important; padding-right: 0.5rem !important; background-color:
            #000000; justify-content: center; align-items: center; border-radius: 9999px !important; width: 13rem;
            margin-top: 10px; margin-left: auto;
        } #share-btn {
            all: initial; color: #ffffff;font-weight: 600; cursor:pointer; font-family: 'IBM Plex Sans', sans-serif;
            margin-left: 0.5rem !important; padding-top: 0.25rem !important; padding-bottom: 0.25rem
            !important;right:0;
        } #share-btn * {
            all: unset;
        } #share-btn-container div:nth-child(-n+2){
            width: auto !important; min-height: 0px !important;
        } #share-btn-container .wrap {
            display: none !important;
        } .gr-form{
            flex: 1 1 50%; border-top-right-radius: 0; border-bottom-right-radius: 0;
        } #prompt-container{
            gap: 0;
        } #generated_id{
            min-height: 700px
        } #setting_id{
          margin-bottom: 12px; text-align: center; font-weight: 900;
        }
"""
iface = gr.Blocks(css=css)

# modified html to only include vital parts
with iface:
    gr.HTML(
        """
            <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; line-height: normal;">
                  AudioLDM Animals: Text-to-Audio Generation with Latent Diffusion Models (hopefully) Fine-Tuned for animal sounds
                </h1>
              </div> <p style="margin-bottom: 10px; font-size: 94%">
                <a href="https://arxiv.org/abs/2301.12503">[Paper]</a> <a href="https://audioldm.github.io/">[Original Project
                page]</a> <a href="https://huggingface.co/docs/diffusers/main/en/api/pipelines/audioldm">[🧨
                Diffusers]</a>
              </p>
            </div>
        """
    )

    with gr.Group():
        with gr.Box():
            textbox = gr.Textbox(
                value="A dog is barking",
                max_lines=1,
                label="Input text",
                info="Your text is important for the audio quality. Please ensure it is descriptive by using more adjectives.",
                elem_id="prompt-in",
            )
            negative_textbox = gr.Textbox(
                value="low quality, average quality",
                max_lines=1,
                label="Negative prompt",
                info="Enter a negative prompt not to guide the audio generation. Selecting appropriate negative prompts can improve the audio quality significantly.",
                elem_id="prompt-in",
            )

            with gr.Accordion("Click to modify detailed configurations", open=False):
                seed = gr.Number(
                    value=45,
                    label="Seed",
                    info="Change this value (any integer number) will lead to a different generation result.",
                )
                duration = gr.Slider(2.5, 10, value=5, step=2.5, label="Duration (seconds)")
                guidance_scale = gr.Slider(
                    0,
                    4,
                    value=2.5,
                    step=0.5,
                    label="Guidance scale",
                    info="Large => better quality and relevancy to text; Small => better diversity",
                )

            outputs = gr.Video(label="Output", elem_id="output-video")
            btn = gr.Button("Submit").style(full_width=True)

        btn.click(
            text2audio,
            inputs=[textbox, negative_textbox, duration, guidance_scale, seed],
            outputs=[outputs],
        )
        
iface.queue(max_size=1).launch(debug=True)