Update app.py

app.py

@@ -4,7 +4,7 @@ import torchaudio
 from audiocraft.models import MusicGen
 from audiocraft.data.audio import audio_write
 import logging
-import os 
+import os
 import uuid
 from torch.cuda.amp import autocast
 import torch
@@ -16,10 +16,16 @@ logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(
 
 logging.info("Carregando o modelo pré-treinado.")
 model = MusicGen.get_pretrained('nateraw/musicgen-songstarter-v0.2')
-model.set_generation_params(duration=500)
+
 
 @spaces.GPU(duration=120)
-def generate_music(description, melody_audio):
+def generate_music(description, melody_audio, duration):
+    # Remove spam da descrição
+    description = clean_text(description)
+
+    # Define a duração da música em segundos
+    model.set_generation_params(duration=int(duration * 1000)) 
+
     with autocast():
         logging.info("Iniciando a geração de música.")
         if description:
@@ -39,7 +45,6 @@ def generate_music(description, melody_audio):
         logging.info(f"Salvando a música gerada com o nome: {filename}")
         path = audio_write(filename, wav[0].cpu().to(torch.float32), model.sample_rate, strategy="loudness", loudness_compressor=True)
         print("Música salva em", path, ".")
-        # 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(path):
@@ -47,30 +52,27 @@ def generate_music(description, melody_audio):
 
         return path
     
+def clean_text(text):
+    """Remove links e caracteres especiais de um texto."""
+    import re
+    text = re.sub(r'http\S+', '', text) # remove links
+    text = re.sub(r'[^a-zA-Z0-9\s]', '', text) # remove caracteres especiais
+    return text
+
 # 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")
+description = gr.Textbox(label="Descrição", placeholder="acústico, guitarra, melodia, trap, ré menor, 90 bpm")
+melody_audio = gr.Audio(label="Melodia de Áudio (opcional)", type="filepath")
+duration = gr.Number(label="Duração (segundos)", value=10)
+output_path = gr.Audio(label="Música Gerada", type="filepath")
 
 gr.Interface(
     fn=generate_music,
-    inputs=[description, melody_audio],
+    inputs=[description, melody_audio, duration],
     outputs=output_path,
     title="MusicGen Demo",
-    description="Generate music using the MusicGen model by Nateraw.\n\n"
-                "Model: musicgen-songstarter-v0.2\n"
-                "Download the model [here](https://huggingface.co/nateraw/musicgen-songstarter-v0.2).\n\n"
-                "musicgen-songstarter-v0.2 is a musicgen-stereo-melody-large fine-tuned on a dataset of melody loops from Nateraw's Splice sample library. "
-                "It's intended to be used to generate song ideas that are useful for music producers. It generates stereo audio in 32khz.\n\n"
-                "Compared to musicgen-songstarter-v0.1, this new version:\n"
-                "- Was trained on 3x more unique, manually-curated samples that Nateraw painstakingly purchased on Splice\n"
-                "- Is twice the size, bumped up from size medium ➡️ large transformer LM\n\n"
-                "If you find this model interesting, please consider:\n"
-                "- Following Nateraw on [GitHub](https://github.com/nateraw)\n"
-                "- Following Nateraw on [Twitter](https://twitter.com/nateraw)\n\n"
-                "Space created by [artificialguybr](https://twitter.com/artificialguybr) on Twitter.",
+    description="Gere música usando o modelo MusicGen by Nateraw.",
     examples=[
-        ["trap, synthesizer, songstarters, dark, G# minor, 140 bpm", "./assets/kalhonaho.mp3"],
-        ["upbeat, electronic, synth, dance, 120 bpm", None]
+        ["trap, sintetizador, iniciantes de música, escuro, Sol# menor, 140 bpm", "./assets/kalhonaho.mp3", 20],
+        ["animado, eletrônico, sintetizador, dança, 120 bpm", None, 30]
     ]
 ).launch()