app.py

import gradio as gr
import scipy
from transformers import pipeline
from transformers import AutoProcessor, MusicgenForConditionalGeneration

processor = AutoProcessor.from_pretrained("facebook/musicgen-small")
model = MusicgenForConditionalGeneration.from_pretrained("facebook/musicgen-small")



def generate(description):
    audio = processor(text=description, padding=True, return_tensors="pt")
    audio_values = model.generate(**audio, max_new_tokens=256)
    sampling_rate = model.config.audio_encoder.sampling_rate
    return scipy.io.wavfile.write("musicgen_out.wav", rate=sampling_rate, data=audio_values[0, 0].numpy())

demo = gr.Interface(
    fn=generate,
    inputs=gr.Textbox(label="Enter Text to Convert to Audio"),
    outputs=gr.Audio(label="Generated Audio"),
    live=True
)

demo.launch()