Spaces:
Running
on
T4
Running
on
T4
File size: 4,562 Bytes
0c98d02 398c97e 0c98d02 18b0529 85d7512 18b0529 c380881 18b0529 6c877a7 18b0529 b0e04c2 02b84d5 b0e04c2 02b84d5 b0e04c2 c0f3731 b0e04c2 fba55a6 b0e04c2 02b84d5 b0e04c2 02b84d5 b0e04c2 fba55a6 b0e04c2 fba55a6 9ca1653 fba55a6 b0e04c2 f841ca8 b0e04c2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 |
import gradio as gr
import torch
from spectro import wav_bytes_from_spectrogram_image
from diffusers import StableDiffusionPipeline
model_id = "riffusion/riffusion-model-v1"
pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16)
pipe = pipe.to("cuda")
def predict(prompt):
spec = pipe(prompt).images[0]
print(spec)
wav = wav_bytes_from_spectrogram_image(spec)
with open("output.wav", "wb") as f:
f.write(wav[0].getbuffer())
return spec, 'output.wav'
title = """
<div style="text-align: center; max-width: 500px; margin: 0 auto;">
<div
style="
display: inline-flex;
align-items: center;
gap: 0.8rem;
font-size: 1.75rem;
margin-bottom: 10px;
"
>
<h1 style="font-weight: 600; margin-bottom: 7px;">
Riffusion real-time music generation
</h1>
</div>
<p style="margin-bottom: 10px;font-size: 94%;font-weight: 100;line-height: 1.5em;">
Describe a musical prompt, generate music by getting a spectrogram image & sound.
</div>
"""
article = """
<p style="font-size: 0.8em;line-height: 1.2em;border: 1px solid #374151;border-radius: 8px;padding: 20px;">
About the model: Riffusion is a latent text-to-image diffusion model capable of generating spectrogram images given any text input. These spectrograms can be converted into audio clips.
<br />—
<br />The Riffusion model was created by fine-tuning the Stable-Diffusion-v1-5 checkpoint.
<br />—
<br />The model is intended for research purposes only. Possible research areas and tasks include
generation of artworks, audio, and use in creative processes, applications in educational or creative tools, research on generative models.
</p>
<div class="footer">
<p>
<a href="https://huggingface.co./riffusion/riffusion-model-v1" target="_blank">Riffusion model</a> by Seth Forsgren and Hayk Martiros -
Demo by 🤗 <a href="https://twitter.com/fffiloni" target="_blank">Sylvain Filoni</a>
</p>
</div>
<p style="text-align: center;font-size: 94%">
Do you need faster results ? You can skip the queue by duplicating this space:
<span style="display: flex;align-items: center;justify-content: center;height: 30px;">
<a href="https://huggingface.co./fffiloni/spectrogram-to-music?duplicate=true"><img src="https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14" alt="Duplicate Space"></a>
</span>
</p>
"""
css = '''
#col-container, #col-container-2 {max-width: 510px; margin-left: auto; margin-right: auto;}
a {text-decoration-line: underline; font-weight: 600;}
div#record_btn > .mt-6 {
margin-top: 0!important;
}
div#record_btn > .mt-6 button {
width: 100%;
height: 40px;
}
.footer {
margin-bottom: 45px;
margin-top: 10px;
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;
}
'''
with gr.Blocks(css=css) as demo:
with gr.Column(elem_id="col-container"):
gr.HTML(title)
prompt_input = gr.Textbox(placeholder="a cat diva singing in a New York jazz club")
send_btn = gr.Button("Get a new spectrogram ! ")
with gr.Column(elem_id="col-container-2"):
spectrogram_output = gr.Image(label="spectrogram image result")
sound_output = gr.Audio(type='filepath', label="spectrogram sound")
gr.HTML(article)
send_btn.click(predict, inputs=[prompt_input], outputs=[spectrogram_output, sound_output])
demo.queue(max_size=250).launch(debug=True)
|