I could not generate the music app not responding

[srirammurty]

I have installed the app as per the installations guidelines. I could able to launch the Gradio interface. When I tried to generate for first time it downloaded all the models required from hugging face. However the app keep running infinitely but not generating any music. I am using Windows 11 with Python 3.11.

**(venv) C:\Sriram\Multi\Music_generation\Application\ACE-Step>acestep --port 7865

• Running on local URL: http://127.0.0.1:7865
• To create a public link, set share=True in launch().
  2025-08-20 22:42:26.629 | WARNING | acestep.pipeline_ace_step:call:1480 - Checkpoint not loaded, loading checkpoint...
  2025-08-20 22:42:26.630 | INFO | acestep.pipeline_ace_step:get_checkpoint_path:178 - Download models from Hugging Face: ACE-Step/ACE-Step-v1-3.5B, cache to: C:\Users\Sriram.cache/ace-step/checkpoints
  Fetching 14 files: 100%|███████████████████████████████████████████████████████████████████████| 14/14 [00:00<?, ?it/s]
  C:\Sriram\Multi\Music_generation\Application\ACE-Step\venv\Lib\site-packages\torch\nn\utils\weight_norm.py:144: FutureWarning: torch.nn.utils.weight_norm is deprecated in favor of torch.nn.utils.parametrizations.weight_norm.
  WeightNorm.apply(module, name, dim)
  2025-08-20 22:42:56.592 | INFO | acestep.pipeline_ace_step:call:1488 - Model loaded in 29.96 seconds.
  2025-08-20 22:42:57.331 | INFO | acestep.pipeline_ace_step:call:1536 - random audio duration: 162.43691869438297
  2025-08-20 22:42:57.332 | INFO | acestep.pipeline_ace_step:text2music_diffusion_process:847 - cfg_type: apg, guidance_scale: 15, omega_scale: 10
  2025-08-20 22:42:57.359 | INFO | acestep.pipeline_ace_step:text2music_diffusion_process:1072 - start_idx: 15, end_idx: 45, num_inference_steps: 60
  0%| | 0/60 [00:00<?, ?it/s]**

After this no more update . Please help to fix this

[srirammurty]

I checked in Microsoft Copilot and found that its because of no GPU usage. Also worked is to update some of the python modules using pip upgdare. Reducing the duration to 20sec , at least the infersteps 20, grading scale 8 the process started on a cpu.

Howeer the warning venv\Lib\site-packages\torch\nn\utils\weight_norm.py:144: FutureWarning: torch.nn.utils.weight_norm is deprecated in favor of torch.nn.utils.parametrizations.weight_norm.
WeightNorm.apply(module, name, dim)

to be updated

[78Alpha]

After waiting a few hours, it appears to be using the CPU, and to an abysmally slow degree. Installing Pytorch GPU manually rectifies it.

And deprecations are normal, something is being deprecated at all times, so until it stops working, it isn't critical at the moment.