Update app.py

app.py

@@ -1,32 +1,60 @@
 import gradio as gr
-import spaces
 import torchaudio
 from audiocraft.models import MusicGen
 from audiocraft.data.audio import audio_write
+import spaces
+import logging
+import os 
+import uuid
+# Configura o logging
+logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
 
+logging.info("Carregando o modelo pré-treinado.")
 model = MusicGen.get_pretrained('nateraw/musicgen-songstarter-v0.2')
-model.set_generation_params(duration=8)  # generate 8 seconds.
+model.set_generation_params(duration=8)
 
-@spaces.GPU(duration=120)  # Specify duration if the function is expected to take more than 60s
-def generate_music(description, audio_file):
-    if audio_file is None:
-        wav = model.generate([description])  # generates 1 sample based on the provided description
+@spaces.GPU(duration=120)
+def generate_music(description, melody_audio):
+    logging.info("Iniciando a geração de música.")
+    if description:
+        description = [description]
+        if melody_audio:
+            logging.info(f"Carregando a melodia de áudio de: {melody_audio}")
+            melody, sr = torchaudio.load(melody_audio)
+            logging.info("Gerando música com descrição e melodia.")
+            wav = model.generate_with_chroma(description, melody[None], sr)
+        else:
+            logging.info("Gerando música apenas com descrição.")
+            wav = model.generate(description)
     else:
-        melody, sr = torchaudio.load(audio_file)
-        wav = model.generate_with_chroma([description], melody[None], sr)  # generates using the melody from the given audio and the provided description
+        logging.info("Gerando música de forma incondicional.")
+        wav = model.generate_unconditional(1)
+    filename = f'{str(uuid.uuid4())}.wav'
+    output_path = os.path.join('./', filename)  # Salva o arquivo no diretório atual
+    logging.info(f"Salvando a música gerada em: {output_path}")
+    audio_write(output_path, wav[0].cpu(), model.sample_rate, strategy="loudness", loudness_compressor=True)
     
-    audio_write('output', wav[0].cpu(), model.sample_rate, strategy="loudness", loudness_compressor=True)
-    return 'output.wav'
+    # Verifica a forma do tensor de áudio e se foi salvo corretamente
+    logging.info(f"A forma do tensor de áudio gerado: {wav[0].shape}")
+    logging.info("Música gerada e salva com sucesso.")
+    if not os.path.exists(output_path):
+        raise ValueError(f'Failed to save audio to {output_path}')
 
-iface = gr.Interface(
-    fn=generate_music,
-    inputs=[
-        gr.Text(label="Description"),
-        gr.Audio(type="filepath", label="Audio File (optional)")
-    ],
-    outputs=gr.Audio(type="filepath"),
-    title="MusicGen",
-    description="Generate music using the MusicGen model. Provide a description and optionally an audio file for melody.",
-)
+    return output_path
+    
+# Define a interface Gradio
+description = gr.Textbox(label="Description", placeholder="acoustic, guitar, melody, trap, d minor, 90 bpm")
+melody_audio = gr.Audio(label="Melody Audio (optional)", type="filepath")
+output_path = gr.Audio(label="Generated Music", type="filepath")
 
-iface.launch()
+gr.Interface(
+    fn=generate_music,
+    inputs=[description, melody_audio],
+    outputs=output_path,
+    title="MusicGen Demo",
+    description="Generate music using the MusicGen model.",
+    examples=[
+        ["trap, synthesizer, songstarters, dark, G# minor, 140 bpm", "./assets/kalhonaho.mp3"],
+        ["upbeat, electronic, synth, dance, 120 bpm", None]
+    ]
+).launch()