app.py

import streamlit as st
import torch
import os
import base64
import torchaudio
import numpy as np
from audiocraft.models import MusicGen

genres = ["Pop", "Rock", "Jazz", "Electronic", "Hip-Hop", "Classical",
          "Lofi", "Chillpop","Country","R&G", "Folk","EDM", "Disco", "House", "Techno",]

@st.cache_resource()
def load_model(model_name):
    model = MusicGen.get_pretrained(model_name)
    return model

def generate_music_tensors(description, duration: int, batch_size=1, models=None):
    outputs = {}
    for model_name, model in models.items():
        model.set_generation_params(
            use_sampling=True,
            top_k=250,
            duration=duration
        )

        with st.spinner(f"Generating Music with {model_name}..."):
            output = model.generate(
                descriptions=description,
                progress=True,
                return_tokens=True
            )
            outputs[model_name] = output

    st.success("Music Generation Complete!")
    return outputs

def save_audio(samples: torch.Tensor, filename):
    sample_rate = 30000
    save_path = "audio_output"
    assert samples.dim() == 2 or samples.dim() == 3

    samples = samples.detach().cpu()
    if samples.dim() == 2:
        samples = samples[None, ...]

    for idx, audio in enumerate(samples):
        audio_path = os.path.join(save_path, f"{filename}_{idx}.wav")
        torchaudio.save(audio_path, audio, sample_rate)
    return audio_path

def get_binary_file_downloader_html(bin_file, file_label='File'):
    with open(bin_file, 'rb') as f:
        data = f.read()
    bin_str = base64.b64encode(data).decode()
    href = f'<a href="data:application/octet-stream;base64,{bin_str}" download="{os.path.basename(bin_file)}">Download {file_label}</a>'
    return href

st.set_page_config(
    page_icon= "musical_note",
    page_title= "Music Gen"
)

def main():
    st.title("🎧 AI Composer 🎧")

    st.subheader("Generate Music")
    st.write("Craft your perfect melody! Fill in the blanks below to create your music masterpiece:")

    bpm = st.number_input("Enter Speed in BPM", min_value=60)
    text_area = st.text_area('Example: 80s rock song with guitar and drums')
    selected_genre = st.selectbox("Select Genre", genres)
    time_slider = st.slider("Select time duration (In Seconds)", 0, 30, 10)
    
    mood = st.selectbox("Select Mood", ["Happy", "Sad", "Angry", "Relaxed", "Energetic"])
    instrument = st.selectbox("Select Instrument", ["Piano", "Guitar", "Flute", "Violin", "Drums"])
    tempo = st.selectbox("Select Tempo", ["Slow", "Moderate", "Fast"])
    melody = st.text_input("Enter Melody or Chord Progression", "e.g., C D:min G:7 C, Twinkle Twinkle Little Star")

    models = {
        'Medium': load_model('facebook/musicgen-medium'),
        'Large': load_model('facebook/musicgen-large'),
        'Large': load_model('facebook/musicgen-melody'),
        'Large': load_model('facebook/musicgen-small'),
                # Add more models here as needed
    }

    if st.button('Let\'s Generate 🎶'):
        st.text('\n\n')
        st.subheader("Generated Music")

        description = f"{text_area} {selected_genre} {bpm} BPM {mood} {instrument} {tempo} {melody}"
        music_outputs = generate_music_tensors(description, time_slider, batch_size=2, models=models)

        for model_name, output in music_outputs.items():
            idx = 0  # Assuming you want to access the first audio file for each model
            audio_filepath = save_audio(output, f'audio_{model_name}_{idx}')
            audio_file = open(audio_filepath, 'rb')
            audio_bytes = audio_file.read()

            st.audio(audio_bytes, format='audio/wav', label=f'{model_name} Model')
            st.markdown(get_binary_file_downloader_html(audio_filepath, f'Audio_{model_name}_{idx}'), unsafe_allow_html=True)

if __name__ == "__main__":
    main()