app.py

"""
Copyright (c) Meta Platforms, Inc. and affiliates.
All rights reserved.

This source code is licensed under the license found in the
LICENSE file in the root directory of this source tree.
"""

from tempfile import NamedTemporaryFile
import torch
import gradio as gr
from scipy.io.wavfile import write

from audiocraft.models import MusicGen

import os
from audiocraft.data.audio import audio_write


MODEL = None

def split_process(audio, chosen_out_track):
    os.makedirs("out", exist_ok=True)
    write('test.wav', audio[0], audio[1])
    os.system("python3 -m demucs.separate -n mdx_extra_q -j 4 test.wav -o out")
    #return "./out/mdx_extra_q/test/vocals.wav","./out/mdx_extra_q/test/bass.wav","./out/mdx_extra_q/test/drums.wav","./out/mdx_extra_q/test/other.wav"
    if chosen_out_track == "vocals":
        return "./out/mdx_extra_q/test/vocals.wav"
    elif chosen_out_track == "bass":
        return "./out/mdx_extra_q/test/bass.wav"
    elif chosen_out_track == "drums":
        return "./out/mdx_extra_q/test/drums.wav"
    elif chosen_out_track == "other":
        return "./out/mdx_extra_q/test/other.wav"
    elif chosen_out_track == "all-in":
        return "test.wav"
    
def load_model(version):
    print("Loading model", version)
    return MusicGen.get_pretrained(version)


def predict(music_prompt, melody, duration, cfg_coef):
    text = music_prompt
    global MODEL
    topk = int(250)
    if MODEL is None or MODEL.name != "melody":
        MODEL = load_model("melody")

    if duration > MODEL.lm.cfg.dataset.segment_duration:
        raise gr.Error("MusicGen currently supports durations of up to 30 seconds!")
    MODEL.set_generation_params(
        use_sampling=True,
        top_k=250,
        top_p=0,
        temperature=1.0,
        cfg_coef=cfg_coef,
        duration=duration,
    )

    if melody:
        sr, melody = melody[0], torch.from_numpy(melody[1]).to(MODEL.device).float().t().unsqueeze(0)
        print(melody.shape)
        if melody.dim() == 2:
            melody = melody[None]
        melody = melody[..., :int(sr * MODEL.lm.cfg.dataset.segment_duration)]
        output = MODEL.generate_with_chroma(
            descriptions=[text],
            melody_wavs=melody,
            melody_sample_rate=sr,
            progress=False
        )
    else:
        output = MODEL.generate(descriptions=[text], progress=False)

    output = output.detach().cpu().float()[0]
    with NamedTemporaryFile("wb", suffix=".wav", delete=False) as file:
        audio_write(file.name, output, MODEL.sample_rate, strategy="loudness", add_suffix=False)
        #waveform_video = gr.make_waveform(file.name)
    return file.name

css="""
#col-container {max-width: 910px; margin-left: auto; margin-right: auto;}
a {text-decoration-line: underline; font-weight: 600;}
"""

with gr.Blocks(css=css) as demo:
    with gr.Column(elem_id="col-container"):
        gr.Markdown(
            """
            # Split Audio Tracks to MusicGen
            Upload an audio file, split audio tracks with Demucs, choose a track as conditional sound for MusicGen, get a remix ! <br/>
            *** Careful, MusicGen model loaded here can only handle up to 30 second audio, please use the audio component gradio feature to edit your audio before conditioning ***
            <br/>
            <br/>
            [![Duplicate this Space](https://huggingface.co/datasets/huggingface/badges/raw/main/duplicate-this-space-sm.svg)](https://huggingface.co/spaces/fffiloni/SplitTrack2MusicGen?duplicate=true) for longer audio, more control and no queue.</p>
            """
        )
            
        with gr.Column():
            uploaded_sound = gr.Audio(type="numpy", label="Input", source="upload")
            with gr.Row():
                chosen_track = gr.Radio(["vocals", "bass", "drums", "other", "all-in"], label="Track", info="Which track from your audio do you want to mashup ?", value="vocals")
            load_sound_btn = gr.Button('Load your chosen track')
            #split_vocals = gr.Audio(type="filepath", label="Vocals")
            #split_bass = gr.Audio(type="filepath", label="Bass")
            #split_drums = gr.Audio(type="filepath", label="Drums")
            #split_others = gr.Audio(type="filepath", label="Other")
        
        with gr.Row():
            music_prompt = gr.Textbox(label="Musical Prompt", info="Describe what kind of music you wish for", interactive=True, placeholder="lofi slow bpm electro chill with organic samples")
            melody = gr.Audio(source="upload", type="numpy", label="Track Condition (from previous step)", interactive=False)
        with gr.Row():
            #model = gr.Radio(["melody", "medium", "small", "large"], label="MusicGen Model", value="melody", interactive=True)
            duration = gr.Slider(minimum=1, maximum=30, value=10, step=1, label="Generated Music Duration", interactive=True)
            cfg_coef = gr.Slider(label="Classifier Free Guidance", minimum=1.0, maximum=10.0, step=0.1, value=3.0, interactive=True)
        with gr.Row():
            submit = gr.Button("Submit")
        #with gr.Row():
        #   topk = gr.Number(label="Top-k", value=250, interactive=True)
        #   topp = gr.Number(label="Top-p", value=0, interactive=True)
        #   temperature = gr.Number(label="Temperature", value=1.0, interactive=True)
        #   cfg_coef = gr.Number(label="Classifier Free Guidance", value=3.0, interactive=True)
        
        output = gr.Audio(label="Generated Music")

        gr.Examples(
            fn=predict,
            examples=[
                [
                    "An 80s driving pop song with heavy drums and synth pads in the background",
                    None,
                    10,
                    3.0
                ],
                [
                    "A cheerful country song with acoustic guitars",
                    None,
                    10,
                    3.0
                ],
                [
                    "90s rock song with electric guitar and heavy drums",
                    None,
                    10,
                    3.0
                ],
                [
                    "a light and cheerly EDM track, with syncopated drums, aery pads, and strong emotions bpm: 130",
                    None,
                    10,
                    3.0
                ],
                [
                    "lofi slow bpm electro chill with organic samples",
                    None,
                    10,
                    3.0
                ],
            ],
            inputs=[music_prompt, melody, duration, cfg_coef],
            outputs=[output]
        )
    load_sound_btn.click(split_process, inputs=[uploaded_sound, chosen_track], outputs=[melody], api_name="splt_trck")
    submit.click(predict, inputs=[music_prompt, melody, duration, cfg_coef], outputs=[output])
    

demo.queue(max_size=32).launch()